[v1][KVCacheManager] Change prefix caching metric from counting blocks to counting tokens

[heheda12345]

If hybrid allocator is introduced, it would be difficult to count the number of hit blocks as blocks will be organized in a complex data structure. To ensure the compatibility with prefix caching metrics, this PR changes the metrics to count the hit tokens. It should give very similar result to count the blocks.

[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.

🚀

[WoosukKwon]

LGTM. @markmc Please let us know if you have any concerns.

[markmc]

LGTM. @markmc Please let us know if you have any concerns.

Short version - LGTM

Longer version - We added queries/hits in #12592 as a replacement for hit_rate in V0. See also here. I guess the hit rate should be the same whether based on blocks or tokens, so I assume there won't be any impact to any existing users.