[V1][Hybrid][Backend] Allow different data types in Hybrid Cache Manager

[bringlein]

Essential Elements of an Effective PR Description Checklist

• The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
• The test plan, such as providing test command.
• The test results, such as pasting the results comparison before and after, or e2e results
• (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.

Purpose

Depends on #21549

Test Plan

Test Result

(Optional) Documentation Update

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

🚀

[gemini-code-assist]

Code Review

This pull request introduces functionality to allow different data types for attention and mamba layers in hybrid models, which is a valuable optimization. However, the current implementation has a few critical issues that need to be addressed. The new configuration options, cache_dtype_attention and cache_dtype_mamba, are used in compute_hash but are commented out in the CacheConfig definition, which will cause an AttributeError. Additionally, there's a NameError in FlashAttentionImpl due to an undefined variable. Finally, a new and seemingly redundant environment variable is introduced in gpu_model_runner.py, creating a confusing configuration path. Addressing these issues will make the feature robust and maintainable.

[mergify]

This pull request has merge conflicts that must be resolved before it can be
merged. Please rebase the PR, @bringlein.

https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks/syncing-a-fork