[TEST]Add 2P1D multi node cases for nightly test

[jiangyunfan1]

What this PR does / why we need it?

This PR adds the 2P1D multi node func/acc/perf test cases, 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@c9461e0

[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 enables and updates a multi-node end-to-end test for the DeepSeek-R1-W8A8 model. The changes include updating benchmark configurations in the YAML file, implementing the test logic in test_multi_node.py to run performance and accuracy tests using aisbench, and a small fix in aisbench.py to handle single test cases. My review focuses on the implementation in test_multi_node.py. I've identified a potential issue with calling blocking code from an asynchronous function, which could impact performance and is against asyncio best practices. I've provided a suggestion to fix this.

[gemini-code-assist]

The run_aisbench_cases function is synchronous and performs blocking I/O operations (e.g., subprocess.Popen and reading from stdout). Calling it directly from an async function like test_multi_node will block the asyncio event loop, which can lead to performance issues and defeats the purpose of using asyncio.

You should run blocking functions in a separate thread to avoid blocking the event loop. You can use asyncio.to_thread for this.

Note: You'll need to add import asyncio at the top of the file.

Suggested change

	run_aisbench_cases(config.model, port, acc_cmd)
	run_aisbench_cases(config.model, port, perf_cmd)
	await asyncio.to_thread(run_aisbench_cases, config.model, port, acc_cmd)
	await asyncio.to_thread(run_aisbench_cases, config.model, port, perf_cmd)