[bugfix] add mlapo test script

[chenjunyi-dev]

What this PR does / why we need it?

This PR mainly does the following things:

1. add mlapo test script
2. support cachemode int8_nzcache with bf16 in mlapo
3. mlapo fit different hidden state dim

Does this PR introduce any user-facing change?

How was this patch tested?

python tests/ops/test_mla_preprocess.py

• 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 introduces several significant improvements. It refactors the mlapo pre-processing logic to support dynamic hidden state dimensions, removing hardcoded values and making the implementation more flexible. It also adds support for int8_nzcache with bfloat16 data type. A key contribution of this PR is the addition of a comprehensive test script for mlapo, which replaces a minimal test with a full test suite. This new suite includes multiple parameter combinations, different cache modes, and two separate "golden" implementations for robust verification. The code changes are clean and the testing is thorough. Overall, this is a high-quality contribution that enhances functionality, flexibility, and test coverage.