feat: Add @langchain/baseten provider package

[Maahir Sachdev (maahir30)]

Adds a new @langchain/baseten package — a LangChain chat model provider for Baseten's OpenAI-compatible inference API. This is a TypeScript port of the official Python langchain-baseten package.

• Introduces ChatBaseten, extending ChatOpenAI and pointing it at Baseten's https://inference.baseten.co/v1 endpoint
• Ports TensorRT-LLM streaming fixes from Python: tool-call chunk normalization (normalizeToolCallChunks) and cumulative usage metadata stripping
• Adds dedicated model URL support (modelUrl) with automatic normalization of /predict, /sync, and /sync/v1 endpoint formats (via normalizeModelUrl)
• Tags all responses with model_provider: "baseten" in response_metadata and ls_provider: "baseten" for LangSmith tracing
• Reasoning content (additional_kwargs.reasoning_content) flows through via the parent @langchain/openai completions converters — verified with pass-through tests
• Defaults streamUsage: true to match the Python package behavior
• Includes 38 unit tests and 3 integration tests (direct invoke, streaming, createDeepAgent end-to-end)

[changeset-bot]

⚠️ No Changeset found

Latest commit: 22aeb49

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR