refact model runner v1

[weiguihua2]

What this PR does / why we need it?

Remove the torchchair logic from model_runner_v1 and put it into torchchair_model_runner

Does this PR introduce any user-facing change?

No

How was this patch tested?

• vLLM version: v0.10.0
• vLLM main: vllm-project/vllm@27e8d1e

[gemini-code-assist]

Code Review

This pull request introduces a significant and well-executed refactoring of the model runner and attention metadata generation logic. The core of the change is the introduction of AscendCommonAttentionMetadata and TorchairCommonAttentionMetadata dataclasses, which centralize the information needed for attention operations. This decouples the attention metadata builders from the main model runner, leading to cleaner code with reduced dependencies and improved modularity. The changes are consistently applied across multiple files, including attention backends and proposers. The refactoring also simplifies some methods and removes redundant code. Overall, this is a high-quality pull request that improves the codebase's structure and maintainability. I have not found any high or critical issues in the proposed changes.

[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.

[github-actions]

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