perf: Update moe gemm (#3239)

<!-- .github/pull_request_template.md -->

## 📌 Description

Update the trtllm-gen cubin to improve mxfp4 moe, DeepSeek-V3 MoE, and
mxfp8 gemm performance.

## 🔍 Related Issues

<!-- Link any related issues here -->

## 🚀 Pull Request Checklist

Thank you for contributing to FlashInfer! Before we review your pull
request, please make sure the following items are complete.

### ✅ Pre-commit Checks

- [x] I have installed `pre-commit` by running `pip install pre-commit`
(or used your preferred method).
- [x] I have installed the hooks with `pre-commit install`.
- [x] I have run the hooks manually with `pre-commit run --all-files`
and fixed any reported issues.

> If you are unsure about how to set up `pre-commit`, see [the
pre-commit documentation](https://pre-commit.com/).

## 🧪 Tests

- [x] Tests have been added or updated as needed.
- [x] All tests are passing (`unittest`, etc.).

## Reviewer Notes

<!-- Optional: anything you'd like reviewers to focus on, concerns, etc.
-->


<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit

* **New Features**
* Improved GPU SM detection to ensure only compatible batched GEMM
configurations are considered.

* **Updates**
* Refined low-latency GEMM kernel selection to use updated tile-size and
dimension variants.
* Adjusted FP8 GEMM kernel selection to new kernel variants for better
coverage.
* Updated artifact references and checksums for TRTLLM GEMM/BMM
binaries.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Signed-off-by: Siyuan Fu <siyuanf@nvidia.com>