[KVCache][ModelRunner] Use vllm InputBatch and Blocktable

[MengqingCao]

What this PR does / why we need it?

This pr change to use InputBatch and BlockTable in vLLM, we just implement the get_supported_kernel_block_sizes in our attention backend, and calculate the block_size and num_block according to the kernel_block_sizes.

How was this patch tested?

test by the exsisting CI

• vLLM version: v0.12.0
• vLLM main: vllm-project/vllm@ad32e3e

[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 is a significant refactoring that removes the custom BlockTable and NPUInputBatch implementations from vllm_ascend. Instead, it now utilizes the InputBatch from the core vLLM library. This change simplifies the codebase by removing duplicated and specialized logic for block and slot management, aligning it more closely with the upstream vLLM implementation. The concept of "hybrid blocks" has been removed and replaced with a more generic mechanism for handling kernel-specific block sizes. The changes are consistently applied across the affected files, including updates to method signatures and call sites. Overall, this is a positive change that improves maintainability and reduces custom code. I have reviewed the changes and found no issues.

[github-actions]

This pull request has conflicts, please resolve those before we can evaluate the pull request.

[MengqingCao]

paged attention branch

[github-actions]

This pull request has conflicts, please resolve those before we can evaluate the pull request.

[github-actions]

This pull request has conflicts, please resolve those before we can evaluate the pull request.

[github-actions]

This pull request has conflicts, please resolve those before we can evaluate the pull request.

[github-actions]

This pull request has conflicts, please resolve those before we can evaluate the pull request.

[wangxiyuan]

no update for long time, close this now. Feel free to reopen if it's still needed.