[Test] add test for prefix cache feature of deepseek

[HuaJiaHeng]

What this PR does / why we need it?

This PR adds a prefix cache case for nightly test for DeepSeek-r1-0528-W8A8 on A3, we need test them daily.

Does this PR introduce any user-facing change?

No

How was this patch tested?

By running the test

• vLLM version: v0.11.0rc3
• vLLM main: vllm-project/vllm@17c540a

[gemini-code-assist]

Code Review

This pull request introduces a new end-to-end test for the prefix cache feature of the DeepSeek model. The test correctly verifies that using a prefix cache improves the Time To First Token (TTFT) by comparing performance on datasets with different prefix-sharing characteristics. My review focuses on improving the maintainability of the new test code. I've identified significant code duplication in how the test cases are executed and have suggested refactoring this logic into a helper function to make the test cleaner and easier to maintain.

[gemini-code-assist]

The test logic for running the benchmark and getting the TTFT is duplicated for aisbench_cases0 and aisbench_cases75. This includes starting the server, warming up, and running the case. This duplication makes the code harder to read and maintain. Any changes to the test setup would need to be applied in two places, increasing the risk of inconsistencies.

To improve this, you can extract the common logic into a helper function. This will make the test cleaner, more readable, and easier to maintain.

    def _run_and_get_ttft(aisbench_case: list) -> float:
        """Helper to start a server, run a benchmark case, and return TTFT."""
        with RemoteOpenAIServer(model,
                                server_args,
                                server_port=port,
                                env_dict=env_dict,
                                auto_port=False):
            run_aisbench_cases(model, port, aisbench_warm_up)
            result = run_aisbench_cases(model, port, aisbench_case)
            return get_TTFT(result)

    TTFT0 = _run_and_get_ttft(aisbench_cases0)
    TTFT75 = _run_and_get_ttft(aisbench_cases75)

[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.

[HuaJiaHeng]

has passed CI.

[jiangyunfan1]

LGTM