[Frontend] Use init_app_state and FrontendArgs from api_server in run_batch

[pooyadavoodi]

Purpose

• Adding support for more features such as tool calling to run_batch.
• This is achieved by using init_app_state and FrontendArgs from vllm/entrypoints/openai/api_server.py.
• The approach taken here also removes some code duplication between api_server and run_batch.
• Due to args conflict over --port between FrontendArgs and the existing run_batch options, we improve the option names from --port and --url to --metrics-port and --metrics-url and provide a backward compatibility guarantee.

Test Plan

Adding a new test for tool calling.

Test Result

$ pytest -v tests/entrypoints/openai/test_run_batch.py
=================================================================================== test session starts ===================================================================================
platform linux -- Python 3.12.9, pytest-9.0.2, pluggy-1.6.0 -- /root/dev/vllm/.venv/bin/python3
cachedir: .pytest_cache
rootdir: /root/dev/vllm
configfile: pyproject.toml
plugins: anyio-4.12.1, asyncio-1.3.0
asyncio: mode=Mode.STRICT, debug=False, asyncio_default_fixture_loop_scope=None, asyncio_default_test_loop_scope=function
collected 8 items

tests/entrypoints/openai/test_run_batch.py::test_empty_file PASSED                                                                                                                  [ 12%]
tests/entrypoints/openai/test_run_batch.py::test_completions PASSED                                                                                                                 [ 25%]
tests/entrypoints/openai/test_run_batch.py::test_completions_invalid_input PASSED                                                                                                   [ 37%]
tests/entrypoints/openai/test_run_batch.py::test_embeddings PASSED                                                                                                                  [ 50%]
tests/entrypoints/openai/test_run_batch.py::test_score[{"custom_id": "request-1", "method": "POST", "url": "/score", "body": {"model": "BAAI/bge-reranker-v2-m3", "queries": "What is the capital of France?", "documents": ["The capital of Brazil is Brasilia.", "The capital of France is Paris."]}}\n{"custom_id": "request-2", "method": "POST", "url": "/v1/score", "body": {"model": "BAAI/bge-reranker-v2-m3", "queries": "What is the capital of France?", "documents": ["The capital of Brazil is Brasilia.", "The capital of France is Paris."]}}] PASSED [ 62%]
tests/entrypoints/openai/test_run_batch.py::test_score[{"custom_id": "request-1", "method": "POST", "url": "/rerank", "body": {"model": "BAAI/bge-reranker-v2-m3", "query": "What is the capital of France?", "documents": ["The capital of Brazil is Brasilia.", "The capital of France is Paris."]}}\n{"custom_id": "request-2", "method": "POST", "url": "/v1/rerank", "body": {"model": "BAAI/bge-reranker-v2-m3", "query": "What is the capital of France?", "documents": ["The capital of Brazil is Brasilia.", "The capital of France is Paris."]}}\n{"custom_id": "request-2", "method": "POST", "url": "/v2/rerank", "body": {"model": "BAAI/bge-reranker-v2-m3", "query": "What is the capital of France?", "documents": ["The capital of Brazil is Brasilia.", "The capital of France is Paris."]}}] PASSED [ 75%]
tests/entrypoints/openai/test_run_batch.py::test_reasoning_parser PASSED                                                                                                            [ 87%]
tests/entrypoints/openai/test_run_batch.py::test_tool_calling PASSED                                                                                                                [100%]

==================================================================================== warnings summary =====================================================================================
<frozen importlib._bootstrap>:488
  <frozen importlib._bootstrap>:488: DeprecationWarning: builtin type SwigPyPacked has no __module__ attribute

<frozen importlib._bootstrap>:488
  <frozen importlib._bootstrap>:488: DeprecationWarning: builtin type SwigPyObject has no __module__ attribute

-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
======================================================================== 8 passed, 2 warnings in 302.24s (0:05:02) ========================================================================

---

Essential Elements of an Effective PR Description Checklist

• The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
• The test plan, such as providing test command.
• The test results, such as pasting the results comparison before and after, or e2e results
• (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.
• (Optional) Release notes update. If your change is user facing, please update the release notes draft in the Google Doc.

[gemini-code-assist]

Code Review

The pull request successfully integrates init_app_state and FrontendArgs from api_server.py into run_batch.py, significantly reducing code duplication and enabling support for new features like tool calling. The changes to argument parsing for metrics, including renaming --port and --url to --metrics-port and --metrics-url respectively, are well-handled with backward compatibility. The addition of a comprehensive test case for tool calling ensures the new functionality works as expected. Overall, the changes improve modularity, maintainability, and extend the capabilities of run_batch.

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

^{Bugbot Autofix is OFF. To automatically fix reported issues with Cloud Agents, enable Autofix in the Cursor dashboard.}

Comment @cursor review or bugbot run to trigger another review on this PR

[DarkLight1337]

We can avoid this by adding a new base class of FrontendArgs, then each subclass can have different definitions of host and port