[feature] Mooncake_connector support pcp/dcp

[wangxiaochao6]

add feature for Mooncake_connector supporting pcp/dcp

• vLLM version: v0.11.0
• vLLM main: vllm-project/vllm@2918c1b

[github-actions]

👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:‌‌

• A PR should do only one thing, smaller PRs enable faster reviews.
• Every PR should include unit tests and end-to-end tests ‌to ensure it works and is not broken by other future PRs.
• Write the commit message by fulfilling the PR description to help reviewer and future developers understand.

If CI fails, you can run linting and testing checks locally according Contributing and Testing.

[gemini-code-assist]

Code Review

This pull request adds support for prefill context parallelism (pcp) and decode context parallelism (dcp) to the Mooncake connector. The changes correctly update parallelism parameters, port calculations, and data structures to handle these new parallelism dimensions. However, I've identified a few critical issues: an incorrect type hint in ReqMeta that will lead to a TypeError, and two separate bugs in the new _get_kv_split_metadata function that will cause incorrect behavior and likely runtime errors when transferring KV caches in pcp/dcp scenarios. These issues need to be addressed to ensure the feature works as intended.

[LCAIZJ]

Noted. I'll monitor this PR for updates.

[wangxiyuan]

@LCAIZJ Please double check this one. Thanks

[LCAIZJ]

@LCAIZJ Please double check this one. Thanks

On it! I'll review and provide feedback shortly.

[wangxiyuan]

@LCAIZJ @liziyu179 Please approve this, then I'll merge it.

[liziyu179]

it's look good to me

[LCAIZJ]

@LCAIZJ @liziyu179 Please approve this, then I'll merge it.

done