🚀 v1.6.0: Multi-Backend Support, PostHog Telemetry, V2 Chat Provider & Model Capability Overrides (#71)

* [Bug]: GPT-5.3-codex Tool Calling Failure
Fixes #54

* [chore] update agent and skills

* fix: Add token counting functionality and update agent tools

This commit introduces the ability to count tokens using LiteLLM's API and updates the tools available to the investigative agent.

✨ Features
- Implement token counting functionality using the LiteLLM API. This allows for more accurate token estimation by leveraging the LiteLLM proxy's capabilities.

🧹 Chores
- Update the investigative agent's tool list to include new capabilities like `vscode/memory`, `browser/openBrowserPage`, `web/githubRepo`, and various `github.vscode-pull-request-github` tools. This enhances the agent's ability to gather information and interact with GitHub.
- Increment the package version from `1.4.3-dev5` to `1.4.3-dev6`.
- Update `llm-response.json` with a comprehensive list of new and updated model configurations, including various Anthropic, OpenAI, and Vertex AI models with their respective parameters and costs. This ensures the system is aware of and can utilize a wider range of LLM providers.

* feat: Enhance diff handling and token counting

### ✨ Features
- Introduce `compactDiff` to reduce the size of Git diffs by removing context lines, making them more suitable for token limits.
- Optimize token counting by introducing a local count and a background refinement process using LiteLLM, with caching for improved responsiveness.
- Add support for `gpt-5-mini` model by correctly stripping unsupported parameters like temperature and top_p.
- Enhance `convertMessages` to properly handle `application/json`, `text/`, and `cache_control` mime types within message parts.

### 🐛 Fixes
- Prevent potential issues with diff truncation by first attempting to compact the diff before hard truncation.

### 🧹 Chores
- Update version to `1.4.3-dev9`.
- Improve model detail display to show context and output token counts more clearly.
- Refine the token count cache with a TTL of 1 minute and a cleanup mechanism for large caches.
- Adjust `DEBOUNCE_MS` for background token count refinement to 300ms.

* feat: Improve model discovery and refresh behavior 🚀

- **✨ Features**
  - Implement `onDidChangeLanguageModelChatInformation` emitter in `LiteLLMProviderBase` to signal changes in model information.
  - Trigger `refreshModelInformation` when the model list changes or when the cache is cleared, ensuring the UI reflects the latest available models.
  - Add relevant keywords to `package.json` for better discoverability of the extension.

- **🧹 Chores**
  - Update the VS Code engine version in `package.json` to `^1.110.0`.
  - Consolidate the `refreshModelInformation` logic within `LiteLLMProviderBase` and remove the deprecated method from `LiteLLMChatProvider`.
  - Expand the tool list for the investigative agent to include a comprehensive set of GitHub-related commands.

* Invstigate/context window usage (#58)

* saving changes, non-functional

* feat: add experimental V2 Chat Provider and V2 message pipeline ✨

**✨ Features**
- Add LiteLLMChatProviderV2 implementing proposed VS Code V2 chat APIs and streaming handling so the extension can use LanguageModelChatMessage2 and Thinking parts.
- Introduce V2 types (v2Types) and a suite of conversion/normalization utilities to map between V2 messages, transport messages, provider messages, and OpenAI payloads.
- Implement V2-aware token utilities (counting and trimming) and emit usage via LanguageModelDataPart fallback when the stream lacks usage metadata.
- Wire an experimental config key (litellm-connector.experimental.v2ApiEnabled) and dynamic provider registration to select V1 or V2 at activation.

**🧪 Tests**
- Add unit tests for the V2 chat provider, stream interpreter, and commit provider behavior, and update existing tests to include requestInitiator metadata for more robust assertions.
- Add tests covering Gemini native format parsing, V2 thinking/data handling, and commit message markdown-stripping behavior.

**🧹 Chores**
- Add stripMarkdownCodeBlocks util and apply it to commit message generation and live input to sanitize code fences for a cleaner UI and outputs.
- Bump package version to 1.4.3-dev18, add enabledApiProposals and update VS Code settings/scripts to include tsc and adjusted dts handling.
- Rename/export emitV2PartsToVSCode (keeping a backward alias) and adjust logging levels (trace/info) across providers for improved diagnostics.

**🐛 Fixes**
- Ensure inline completion logging reflects informational state and sanitize streamed commit messages by stripping markdown code blocks to prevent backticks in final messages.

* fixing ignores...

* chore: tighten .gitignore globs for VS Code type files 🧹

- **🧹 Chores**
  - Replace **/*vscode.d.ts and **/*vscode.proposed.* with **/vscode.d.ts and **/vscode.proposed.* to avoid over-matching. Only files named exactly vscode.d.ts or vscode.proposed.* will be ignored.

* chore: update CI workflows, actions, and dev dependencies 🔧

**🧹 Chores**
- Update Dependabot to ignore major updates for @types/node beyond v22 to avoid major-version PRs.
- Use ${{ github.workspace }}/... paths for Codecov uploads and point test results to ${{ github.workspace }}/test-results/test-results.xml so CI reliably finds coverage and test artifacts.
- Bump actions/upload-artifact to v7 in CI and actions/download-artifact to v8 in release to use newer action releases.
- Bump devDependencies: upgrade @eslint/js and eslint to ^10.0.0, update @types/mocha and @types/node to patch versions, and remove all-contributors-cli.

* fix: include original errors as cause when rethrowing 🐛

**🐛 Fixes**
- Include original error as the cause when rethrowing in LiteLLMClient and LiteLLMChatProvider so cancellation and API errors preserve context and stack traces.
- Attach retryErr as the cause when throwing retryErrorMessage to retain the original retry failure details.
- Include the original error as cause when emitting the final error to improve diagnostics and logging.

**🧹 Chores**
- Bump package.json version to 1.4.3-dev20.
- Add a comment in convertMessages catch to document the fallback behavior when JSON.stringify fails.

* ver bump in prep for release.  [skip ci]

* chore: 🚀 Finalize project structure and improve streaming/observability

**🧹 Chores**
- Update `.devcontainer/devcontainer.json` to include additional VS Code extensions for better development tooling.
- Refresh agent definitions (`Project Planner`, `Changelog Manager`, `Investigative Agent`) and add new skills (`generate-github-bug-report`, `plan-creation-and-refinement`, `ts-no-any`) to improve automation standards.
- Update CI/CD workflows and `package.json` with new versions and setting deprecations.
- Implement comprehensive observability layer (`AuditTrail`, `HookSystem`, `StructuredLogger`) to improve debugging for the v2 provider.
- Update `CHANGELOG.md` and repository `README.md` to reflect current project features and standards.

**🐛 Fixes**
- Refactor `decodeSSE` in `src/adapters/sse/sseDecoder.ts` to correctly handle multiline SSE payloads and improve stream stability.
- Fix `liteLLMStreamInterpreter` to properly deduplicate tool calls and prevent corruption of JSON arguments.
- Resolve tool call normalization issues (ensure IDs are ≤ 40 chars) for compatibility with strict providers.
- Add regression tests in `src/adapters/sse/test/sseDecoder.test.ts` and `src/adapters/streaming/test/liteLLMStreamInterpreter.test.ts`.

**✨ Features**
- Add `litellm-connector.setLogLevel` command to allow runtime adjustments to logging verbosity.
- Add `sendDefaultParameters` config setting for greater control over model hyperparameters.
- Introduce `in-flight deduplication` for model discovery to prevent excessive API lookups.

* refactor: 🛠️ improve logging architecture and cleanup

- **✨ Features**
  - Rename structured logger output channel to "LiteLLM Structured" to prevent collision with legacy logs.
- **🧹 Chores**
  - Remove unused `setLogLevel` command and registration logic.
  - Bump project version to `1.5.0-dev2` and apply minor formatting to `package.json`.
- **🧪 Tests**
  - Add unit test verifying that the structured logger initializes with the correct, distinct output channel name.

* feat: release version 1.5.0 and improve provider registration 🚀

- **🚀 Features**
  - Improve model discovery, refresh behavior, and token counting functionality.
  - Enhance agent tools and diff handling for better model interaction.
- **🛠️ Bug Fixes**
  - Resolve tool calling failures for GPT-5.3-codex and fix 400 errors for Gemini and o1-series models.
- **🏠 Internal**
  - Refactor logging architecture and rename "LiteLLM V2" output channel to "LiteLLM Structured".
  - Implement explicit `LanguageModelChatProvider` registration in the extension activation flow.
- **📚 Docs**
  - Update README and marketplace documentation with improved quick-start instructions and pro-tips for V2 Chat features.
- **🧹 Chores**
  - Bump package version to 1.5.0.

* feat: 🚀 add multi-backend support and strict TypeScript guidelines

- **✨ Features**
  - Implement multi-backend orchestration to support model aggregation and namespacing across multiple LiteLLM instances.
  - Introduce a comprehensive TypeScript "No-Any" policy to enforce strict type safety and eliminate technical debt.
- **📦 Documentation**
  - Add the full LiteLLM OpenAPI specification to serve as a foundational reference for proxy API integrations.
  - Update the changelog and agent documentation to reflect version 1.6.0-dev milestones and architectural standards.
- **🧹 Chores**
  - Upgrade Codecov CI action to v6 and remove unused extensions from the development container configuration.
  - Implement a post-session validation hook to automate execution of verification scripts.

* Integrate PostHog telemetry and enhance project planning capabilities (#70)

* feat: 📊 integrate PostHog telemetry and add plan generation standards

**✨ Features**
- Introduce a comprehensive `TelemetryService` using PostHog to track extension activation, feature usage, and performance metrics across Node.js and Web environments.
- Instrument chat providers, inline completions, and configuration commands to capture detailed event data, including token usage and error types.
- Implement an observability hook system via `PostHogHook` to automatically report request completion and message trimming events.

**📝 Documentation**
- Add `plan-generation.instructions.md` to define strict standards for generating autonomous, code-explicit implementation plans.

**🧹 Chores**
- Update project dependencies to include `posthog-js` and `posthog-node` while bumping the version to `1.6.0-dev4`.
- Configure `esbuild` and `tsconfig` path aliases to support platform-specific telemetry adapters.

**🧪 Testing**
- Add unit tests for `TelemetryService` to verify correct event capturing and settings respect.

* feat: 🚀 Upgrade planning agent and integrate PostHog error tracking

- **✨ Features**
  - Refactor `Project Planner` agent to adopt standardized `plan-generation.instructions.md` and switch from `.plan.md` to `.prompt.md` files.
  - Implement full PostHog error tracking with automated source map uploads via `npm run posthog:sourcemaps`.
  - Add configuration schema for multiple LiteLLM backends.
  - Add `CHANGELOG.md` support in VS Code packaging.

- **🧹 Chores**
  - Update `esbuild` to support external source map generation and improved build logging.
  - Update project version to `1.6.0-dev7`.
  - Enhance `TelemetryService` to support exception capture, user identification, and feature flag checks.
  - Update `Logger` to automatically report errors to `TelemetryService`.
  - Add unit tests for telemetry adapters.

* feat: add centralized telemetry and update pre-release version 🚀

- **✨ Features**
  - Initialize centralized `TelemetryService` and wire observability hooks into core request paths for consistent tracking.
  - Enrich telemetry data with dimensions like OS, extension version, and UI context to improve event analysis.
- **🐛 Fixes**
  - Update `post-session-validation.sh` to handle stdin gracefully only when input is provided.
- **🧹 Chores**
  - Bump version to `1.6.0-dev8` and remove outdated `baseUrl` from `tsconfig.json`.

* feat: 🛰️ Implement comprehensive telemetry and add project planning agent

- **✨ Features**
  - Add `project-planner` agent prompt to streamline implementation planning for new tasks.
  - Implement feature-usage reporting across core commands and providers, including chat, completions, and model picking.
  - Add `ConfigManager` integration to capture feature toggle snapshots on startup and report toggle changes in real-time.
  - Wire command execution tracking into model management and reset commands for better usage visibility.
  - Enhance `TelemetryService` with exception capture including caller metadata for better error traceability.

- **🧹 Chores**
  - Update extension version to `1.6.0-dev10` and consolidate development changes.
  - Remove redundant PostHog documentation from README as telemetry is now fully integrated.
  - Cleanup unused code paths and fix minor shell script status codes.

- **🧪 Tests**
  - Add unit tests for `ConfigManager` toggle reporting and verify telemetry event payloads.
  - Expand test coverage for `TelemetryService` to validate snapshot and feature usage event tracking.

* feat: 🚀 improve telemetry identification and bump version

- **✨ Features**
  - Use `vscode.env.sessionId` as the primary distinct ID for better crash reporting correlation.
  - Standardize `extension_version` property usage across all telemetry events and identity calls.
- **🧪 Tests**
  - Add test case to verify that `identify` calls correctly capture the session ID and extension version.
- **🧹 Chores**
  - Bump package version to 1.6.0-dev11.

* test: 🧪 update telemetry unit tests to use sessionId for distinctId

- **🧪 Tests**
  - Update `TelemetryService` tests to use `vscode.env.sessionId` instead of `machineId` as the source for `distinctId`.
  - Remove redundant `identify` test case to clean up the test suite.

* chore: 🛠️ Add implementation prompt, update version, and expand test coverage

**✨ Features**
- Add `implement-plan.prompt.md` to support standardized plan implementation workflows.

**🧹 Chores**
- Bump extension version to `1.6.0-dev12`.

**🧪 Tests**
- Add tests to `liteLLMStreamInterpreter` for tool call deduplication, response format edge cases, and Gemini native format.
- Add connectivity, token counting, and chat edge case tests for `LiteLLMClient`.
- Expand `ResponsesAdapter` testing for combined text/tool calls, tool name synthesis, and complex message content handling.
- Add `ResponsesClient` tests for anonymous tool calls, delta variations, and usage reporting safety.
- Add comprehensive token budget and V2 message normalization unit tests in `tokenUtils` and `utils`.
- Add robust tests for `manageConfig` commands and `modelPicker` error scenarios.
- Improve test coverage for telemetry adapters and base provider functionality, including error handling and model tagging.
- Ensure extension activation properly handles edge cases like missing metadata or registration failures.

* feat: 🚀 enhance provider compatibility and robust request validation

**✨ Features**
- Support native streaming and vision modalities in `LiteLLMProviderBase` by adding relevant tags to the model info.
- Improve `ResponsesClient` reliability by adding error handling to `DataPart` usage reports.
- Update `stripMarkdownCodeBlocks` to specifically target triple backticks, preserving single backticks within text.

**🐛 Fixes**
- Enhance `validateRequest` to strictly check for empty message content across all request messages.
- Allow `assistant` role string in `trimV2MessagesForBudget` for broader compatibility.
- Ensure `isContinuation` logic properly breaks the loop to prevent message processing errors.

**🧹 Chores**
- Bump project version to `1.6.0-dev13`.
- Update unit tests to reflect updated logic for progress reporting, message budget trimming, and validation.

* refactor: 📉 reduce telemetry verbosity

- **🧹 Chores**
  - Comment out excessive telemetry tracking across providers to reduce noise in logs.
  - Temporarily disable successful request telemetry in `PostHogHook` and `LiteLLMTelemetry` to streamline output.
  - Bump project version to `1.6.0-dev14`.

* test: 🧪 update fetch stubbing in LiteLLM client tests

- **🧪 Tests**
  - Refactor fetch stubbing to use `returns` with resolved promises for better control over mock behavior.
  - Ensure consistent async handling during rate limit retry testing.

* feat: 🚀 Enhance telemetry system and tracking

- **✨ Features**
  - Add documentation and automation prompts to standardize telemetry implementation across the codebase.
  - Implement aggregated feature usage tracking in `TelemetryService` to reduce payload frequency by batching events every 15 minutes.
  - Add specific tracking for model/provider usage and feature adoption to gain insights into user behavior.
- **🐛 Fixes**
  - Refactor `captureFeatureUsed` to support aggregation, ensuring performance is not degraded by high-volume events.
- **🧹 Chores**
  - Increment project version to `1.6.0-dev15`.
  - Update `extension.ts` to trigger initial feature adoption events on activation.
  - Update `LiteLLMChatProvider` and `LiteLLMProviderBase` to capture model usage metrics during requests and discovery.
  - Add extensive unit tests for aggregated telemetry and new tracking methods in `telemetryService.test.ts`.

* feat: 🚀 Add changelog automation and improve telemetry observability

- **✨ Features**
  - Add a new automation prompt to generate `CHANGELOG.md` updates from completed plan files.
- **🚀 Enhancements**
  - Propagate `request_id` across telemetry events to allow for better request lifecycle correlation in PostHog.
  - Improve `TelemetryService` initialization to prioritize `machineId` over `sessionId` for more stable user identification.
- **🐛 Fixes**
  - Ensure commit-message generation errors are surfaced in logs and telemetry instead of being silently swallowed.
  - Fix test suite cleanup by introducing a Sinon sandbox to prevent cross-test interference.
- **🧪 Testing**
  - Add comprehensive regression tests to verify `request_id` propagation and error reporting in telemetry.

* feat: 🚀 Add support for multiple LiteLLM instances

### ✨ Features
- Implement support for multiple backend and LiteLLM instances in the legacy method.
- Integrate telemetry for non-identifiable data collection and reporting.

### 🛠️ Bug Fixes
- ⚙️ Fix undefined silent flag, resolving configuration issues for LiteLLM instances.
- 🛠️ Address nuke command failures that leave stale models during configuration resets.
- 🛠️ Stabilize telemetry ID alignment and re-scope unhandled exception management.

### 📊 Telemetry & Observability
- 📈 Add feature usage reporting for chat, completions, and model picker events.

* feat: 🎉 Add Model Capability Overrides configuration to improve VS Code support

# ✨ Features
- Introduce configuration for manual capability overrides (toolCalling, imageInput) to address inaccuracies in LiteLLM's auto-detection. Configure using `litellm-connector.modelCapabilitiesOverrides`.

# 🛠️ Bug Fixes
- Ensure commit-message generation failure visibility in both local logs and telemetry.

* test: add unit tests for PostHogHook functionality 🎉

**🧪 Tests**
- Introduce unit tests for the `PostHogHook` class to ensure correct registration of hooks and functionality for success and error paths.
- Validate `captureRequestCompleted` and `captureRequestFailed` calls based on the metadata passed in the context.
- Ensure that `captureTrimExecuted` is called correctly based on the trim conditions and defaults to zero when metadata is missing.
- Include tests for the safe disposal of the hook to prevent errors on multiple calls to dispose.

* chore: update CI workflows and coverage thresholds 🎉

**📦 CI Changes**
- Removed deprecated branch `geth-dev` and its patterns from workflows to streamline CI setup.
- Retained only necessary branches to trigger CI processes.

**🔧 Coverage Updates**
- Increased coverage threshold from 10% to 20% in `codecov.yml` to improve code quality standards.

* release: bump version to 1.6.0-rc1 🎉

### 🚧 Development snapshot
- Consolidate the current 1.6 pre-release work on rc1.

### 🎊 New Features
- Added telemetry and non-identifiable data collection & reporting.

* feat: select repository by SCM rootUri when generating commit messages ✨

**✨ Features**
- Use the SCM context's rootUri to choose which repository GitUtils.getStagedDiff reads so the staged diff comes from the correct workspace.
- Prefer the repository matching the SCM context when locating the SCM input box, falling back to the first repository for backward compatibility.

**🧹 Chores**
- Add GitUtils.findRepositoryByRootUri and update getStagedDiff to accept an optional rootUri, with safer internalRepo typing and fallback behavior.
- Expose rootUri on the Repository interface to enable reliable repo matching by workspace path.

**🧪 Tests**
- Add unit tests confirming getStagedDiff uses the correct repository and that generated commit messages are written to the matching repo's input box.
- Add tests ensuring getStagedDiff falls back to the first repo when no rootUri is provided and that compactDiff removes excess context lines.

* updating change log and version

* test: add shared test mocks and stabilize unit tests ✨

**🧪 Tests**
- Add comprehensive messageNormalization.test.ts to explicitly test role mapping and normalization logic.
- Consolidate and move the showModels quick-pick clipboard regression into manageConfig tests and refine assertions to reduce brittleness.

**🧹 Chores**
- Introduce createMockSecrets and createMockOutputChannel helpers in src/test/utils/testMocks.ts to DRY repeated secret and output-channel mocks.
- Replace many inline secret/output-channel mocks across tests with the new helpers for consistency and clearer test setup.

**🐛 Fixes**
- Enhance client test response mock to emulate headers.get("retry-after"), statusText, and clone, improving fidelity of fetch-response tests.
- Tweak various stubs and assertions (showQuickPick item shapes, showInputBox behavior, withProgress/checkConnection flows, and assert.ok usage) to make tests more reliable.

* test: add unit tests for testMocks utilities 🧪

**🧪 Tests**
- Add tests for TEST_USER_AGENT to verify it is a non-empty string and suitable for user-agent usage.
- Add comprehensive createMockSecrets tests covering default values, custom baseUrl/apiKey overrides, object and Map entries, store/delete behavior, keys enumeration, and onDidChange disposable handling.
- Add createMockModel tests to validate sensible defaults, full overrides, and that partial overrides preserve unspecified fields.
- Add createMockOutputChannel tests to assert presence of log/UI/content methods and that all methods are safe no-ops (do not throw).

* chore: unify build/CI tooling with zx script, add version utils & tests ✨

**✨ Features**
- Add a unified zx-based build script (scripts/build.mjs) to run build, dev, ci, package, test, validate, upgrade, and version commands from a single entrypoint.
- Expose many npm script aliases that delegate to the zx build script to simplify developer workflows.

**🐛 Fixes**
- Include vscode.LanguageModelThinkingPart handling in LiteLLMChatProvider so thinking-part tokens are accumulated and total output tokens are counted correctly.

**🧹 Chores**
- Replace separate CI steps with a single GitHub Actions step that runs zx scripts/build.mjs ci and set VSCODE_TEST_RESULTS_DIR for test artifacts.
- Update docker-build workflow triggers to include package.json, package-lock.json, and scripts/build.mjs.
- Bump package version to 1.6.0-rc3 and consolidate/replace many npm scripts with zx-based commands.
- Upgrade dev deps (typescript -> ^6.0.3) and add zx (^8.8.5), and adjust posthog sourcemap packaging to use the new build flow.

**🧪 Tests**
- Add version utilities (scripts/versionUtils.mjs) and unit tests (scripts/test/build-version.test.mjs) to validate version parsing/bumping logic.

* feat: disable PostHog telemetry in CI/mock environments 🛑

**✨ Features**
- Add an offlineMode flag to Node and Web PostHog adapters that disables initialization when running in CI or when POSTHOG_MOCK is set, preventing outgoing telemetry.

**🐛 Fixes**
- Guard flush and shutdown to no-op when the adapter is disabled or the client is missing to avoid calling methods on undefined objects.

**🧹 Chores**
- Enable POSTHOG_MOCK in the CI workflow and in package.json test scripts so test runs do not send telemetry.

**✅ Tests**
- Update tests to save/restore CI/POSTHOG_MOCK env vars, stub additional PostHog methods, and add assertions that reloadFeatureFlags is a no-op when disabled and that flush/shutdown delegate when enabled.

* fix: bad ci.yml command

* refactor: centralize package metadata handling in build script ♻️

**♻️ Refactors**
- Add withPackageMetadata helper to manage preview and conditionally strip enabledApiProposals, run an action, and reliably restore package.json afterwards.
- Replace the previous ad-hoc strip/restore logic with withPackageMetadata calls in the marketplace, package, and package:dev build flows to consolidate behavior.

**🧹 Chores**
- Remove legacy stripProposals and restoreProposals functions from scripts/build.mjs.
- Move enabledApiProposals entry in package.json to the end of the file.

* feat: add package-marketplace script and model reasoning/pdf capability support ✨

**✨ Features**
- Add scripts/package-marketplace.mjs to produce a marketplace-ready VSIX by swapping README and adjusting package metadata during packaging.
- Expose ModelCapabilityOverride fields (reasoning, pdfInput) and derive supportsReasoning/supportsPdf to add "reasoning" and "pdf" tags for models.

**🧹 Chores**
- Remove legacy scripts/build.mjs and the old scripts/test/build-version.test.mjs to simplify build tooling.
- Update CI to run tests with coverage and invoke the new package:marketplace step, and remove package.json/scripts/build.mjs from docker-build inputs.
- Add "dennisvhest.npm-browser" to the devcontainer extensions.

**📚 Docs**
- Add .github/prompts/update-readmes-from-repo-state.prompt.md to help update README files from repo state.
- Expand .github/instructions/plan-generation.instructions.md with a Deliverables section clarifying plan file placement and update behavior.

**🐛 Fixes**
- Prevent blocking on hook input by draining stdin in scripts/post-session-validation.sh.

**🧪 Tests**
- Add and update modelCapabilities tests to verify reasoning and pdf tags are added from both derived capabilities and overrides.

* updated package file

* feat: add multi-backend & V2 chat support, telemetry, model capability overrides ✨

**✨ Features**
- Add multi-backend aggregation and multi-repo commit generation so multiple LiteLLM proxies and multi-repo workspaces are supported.
- Introduce experimental V2 Chat Provider, telemetry & observability, model capability overrides, and advanced token counting to improve provider features and diagnostics.
- Add new configuration settings (backends, baseUrl, modelOverrides, modelCapabilitiesOverrides, emitUsageData, etc.) to expose the new features and behaviors.

**📝 Docs**
- Replace VS Code Marketplace badges with Open VSX badges and update README(s) to require VS Code 1.110+ and reflect new features and settings.
- Replace legacy settings docs with a structured settings table and update commands, license, and feedback/contribution sections for clarity and accuracy.
- Tweak wording/formatting (e.g., apostrophes, deprecation notes) across README files for consistency.

**🧪 Tests**
- Update integration error handling tests to stub provider._configManager.isConfigured and getConfig, supplying a local proxy URL and inactivityTimeout for test scenarios.