feat(evaluation): add tool parameter matching

[gioboa]

Please ensure you have read the contribution guide before creating a pull request.

Link to Issue or Description of Change

1. Link to an existing issue:

• Related: Proposal: Add built-in tool_parameter_match evaluator #5643

Problem:

The existing trajectory metric scores a tool call as entirely correct or incorrect. It cannot show partial correctness when only some expected tool arguments match.

Solution:

Add an exact tool_parameter_match metric that averages matches across expected argument keys. Register the metric and cover matching, alignment, empty-input, and partial-score cases with unit tests. This is intentionally a small MVP;advanced matching strategies proposed in #5643 are outside this PR.

Testing Plan

Unit Tests:

• I have added or updated unit tests for my change.
• All relevant unit tests pass locally.

Results:

• Evaluation unit suite: 843 passed
• Final focused suite: 40 passed
• Ruff, isort, Pyink, mypy, compliance, and codespell checks passed.

Manual End-to-End (E2E) Tests:

Compared an expected search(city="Rome", rooms=1) call with an actual
search(city="Rome", rooms=2) call.

Before:

tool_trajectory_avg_score=0.0

After:

tool_parameter_match=0.5, status=PASSED

Checklist

• I have read the CONTRIBUTING.md document.
• I have performed a self-review of my own code.
• I have commented my code, particularly in hard-to-understand areas.
• I have added tests that prove my fix is effective or that my feature works.
• New and existing relevant unit tests pass locally with my changes.
• I have manually tested my changes end-to-end.
• No dependent downstream changes are required for this code change.

Additional context

This PR implements exact argument matching only. Match modes, configurable per-argument strategies, and numeric tolerances can be added separately after maintainer feedback. User-facing documentation will be proposed in the adk-docs repository once the API is approved.