Enable compression of GPU device binaries (#2949)

- This significantly reduces the size of flashinfer-jit-cache on disk
(e.g., 6.6 GB -> 1.5 GB).
- Note that enabling compression is standard for CUDA libraries (e.g.,
CUDA Toolkit libraries use it).

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

## 📌 Description

<!-- What does this PR do? Briefly describe the changes and why they’re
needed. -->

## 🔍 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

- [ ] I have installed `pre-commit` by running `pip install pre-commit`
(or used your preferred method).
- [ ] I have installed the hooks with `pre-commit install`.
- [ ] 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

- [ ] Tests have been added or updated as needed.
- [ ] 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

* **Chores**
* Updated build settings so generated device binaries are now
compressed, reducing their on-disk size. This change is transparent to
users and preserves runtime behavior and existing public interfaces. No
functional or API changes were made.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Co-authored-by: Zihao Ye <expye@outlook.com>