Preserve dict output Field metadata

[Whning0513]

Summary

• preserve top-level Annotated/Field metadata for dict[str, T] tool return types
• pass the original return annotation into the dict RootModel path
• add a regression test covering Field(description=...) on dict outputs

Testing

• PYTHONPATH=src:src/mcp-types python -m pytest -q tests/server/mcpserver/test_func_metadata.py -k structured_output_dict_str_types
• PYTHONPATH=src:src/mcp-types python -m pytest -q tests/server/mcpserver/test_func_metadata.py

Fixes #2935

[Copilot]

Pull request overview

This PR fixes MCPServer tool output-schema generation so that dict[str, T] return types preserve top-level Annotated[...] / Field(...) metadata (e.g., description) in the generated JSON schema, aligning dict outputs with other structured output return types.

Changes:

• Pass the original (potentially Annotated) return annotation into the dict[str, T] RootModel creation path.
• Add a regression test asserting that Field(description=...) on Annotated[dict[str, int], ...] appears in meta.output_schema.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
src/mcp/server/mcpserver/utilities/func_metadata.py	Ensures dict[str, T] structured output schema generation uses the original return annotation so Annotated/Field metadata is preserved.
tests/server/mcpserver/test_func_metadata.py	Adds a regression assertion that description is retained for Annotated[dict[str, T], Field(...)] outputs.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[cubic-dev-ai]

No issues found across 2 files

<sub>Tip: cubic could auto-approve low-risk PRs like this, if it thinks it's safe to merge. Learn more

Re-trigger cubic</sub>