add mla_preprocess kernel

[kiscad]

What this PR does / why we need it?

• Adds the mla_preprocess custom kernel to provide an optimized pre-processing operator for Multi-head Latent Attention (MLA) on Ascend NPUs.
• Wires the new kernel into the C++ extension pipeline so vLLM can invoke it directly, cutting Python-side tensor shuffling and memory copies that previously bottlenecked MLA compilation paths.

Does this PR introduce any user-facing change?

• No. The change only introduces a low-level kernel; public APIs and inference behavior remain unchanged.

How was this patch tested?

• Dedicated Ascend kernels are not covered by our CI yet, so no extra automated tests were added. Future MLA-focused regression runs will cover this path.

• vLLM version: v0.11.0rc3

• vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.0

[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 introduces a new mla_preprocess kernel, which involves significant changes across the host-side tiling logic, kernel implementation, and PyTorch bindings. The implementation is complex and adds a substantial amount of new code. My review focuses on critical aspects of correctness, performance, and thread safety. I've identified a critical race condition in the host-side tiling logic that needs to be addressed to prevent data corruption in multi-threaded environments. Additionally, there are opportunities to improve kernel performance by optimizing memory copy operations and to enhance correctness and performance on the host by replacing floating-point calculations with integer arithmetic for tiling parameters.

[kiscad]

Hi @Yikun @wangxiyuan 👋
This PR is ready for review. Please take a look when you have time.
Let me know if any clarification is needed. Thanks!

[Yikun]

@kiscad

1. Please make sure 310P image build test passed.
2. Would you mind add a python e2e test: https://github.com/vllm-project/vllm-ascend/tree/main/tests/e2e/singlecard/ops to make sure mlapo ops works as expected? (I guess you can move a simple test from atb repo)

[ApsarasX]

So how should we use this kernel in the modeling file?

[kiscad]

So how should we use this kernel in the modeling file?

Yes, this kernel will be used with MLA and DSA (deepseek sparse attention) for better performance, and the adaptation of model implements will be submitted later.