chore: Update ToolHive API models

[stackloke2e]

Summary

This PR updates the ToolHive API models generated from the latest OpenAPI specification.

Changes

• Updated Pydantic models in src/mcp_optimizer/toolhive/api_models/

Notes

• This PR was automatically generated by the update-thv-models workflow
• The models are generated using datamodel-codegen from ToolHive's OpenAPI endpoint

🤖 Generated with GitHub Actions

[claude]

PR Review

Summary

Auto-generated update of ToolHive API models from OpenAPI spec. The changes are primarily modernizing Python type hints from Optional[T] to T | None and from typing imports to native Python 3.10+ union types.

✅ Positives

• Modern type hints: Correctly uses native Python union types (str | None) instead of Optional from typing module
• Consistent changes: All 17 files updated uniformly following the same pattern
• New field added: scopes field added to TokenValidatorConfig in auth.py:59-62 for OAuth scope handling (RFC 9728)
• Removed deprecated field: tool_type removed from Workload model in core.py (was always "mcp")
• Auto-generated: Uses datamodel-codegen, reducing manual errors

🔍 Observations

1. Breaking change: Removed tool_type field from Workload model (core.py). Any code depending on this field will break
2. Type modernization: All type hints now use PEP 604 union syntax (X | None), requiring Python 3.10+
3. Timestamp update: Generation timestamp changed from 2025-12-02 to 2026-01-08

⚠️ Recommendations

1. Version check: Ensure pyproject.toml requires Python 3.10+ (needed for X | None syntax)
2. Breaking change: Document the removal of tool_type field in release notes if this is user-facing
3. Consistency: Consider updating /src/mcp_optimizer/config.py which still uses from typing import Optional for consistency

🎯 Verdict

LGTM with minor note - Clean auto-generated update. Main concern is the breaking change with tool_type removal - verify this does not impact existing usage.