feat(evaluation): add option to save eval results to CSV

[vaibhav-patel]

Summary

Adds an optional, opt-in way to persist AgentEvaluator results to a CSV file, as requested in #2652. A new output_file parameter is added to both AgentEvaluator.evaluate and AgentEvaluator.evaluate_eval_set. When provided, per-invocation results for every metric (both passing and failing) are written to the given path; when omitted (the default), behavior is unchanged.

Motivation

Running agent evals from pytest currently only prints detailed tables (and only for failing metrics). Users want to save results to disk for later inspection/tracking. This implements the CSV suggestion from #2652, adapted to the current evaluation architecture (the codebase has changed substantially since the issue was filed).

What changed

• New output_file: Optional[str] = None on AgentEvaluator.evaluate and evaluate_eval_set.
• Results are flattened to one row per metric per invocation. Columns: eval_set_id, eval_id, metric_name, threshold, score, eval_status, prompt, expected_response, actual_response, expected_tool_calls, actual_tool_calls.
• Parent directory is created if needed; rows are appended (header written once) so evaluating a directory of .test.json files accumulates results in a single file.
• Reuses existing content/tool-call formatting helpers; CSV writing uses pandas, already declared in the eval optional dependencies.

Backward compatibility

Fully backward-compatible. The feature is disabled unless output_file is set.

Testing

Added tests/unittests/evaluation/test_agent_evaluator.py covering row flattening, the missing-expected-invocation case, single-file writing with directory creation, and append-without-duplicate-header. All pass; the full tests/unittests/evaluation/ suite still collects cleanly.

Fixes #2652.

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[vaibhav-patel]

@googlebot I signed it!

[rohityan]

Hi @vaibhav-patel , Thank you for your contribution! We appreciate you taking the time to submit this pull request. Your PR has been received by the team and is currently under review. We will provide feedback as soon as we have an update to share.

[rohityan]

Hi @ankursharmas , can you please review this.

[adk-bot]

Thank you @vaibhav-patel for your contribution! 🎉

Your changes have been successfully imported and merged via Copybara in commit 3bbc8ed.

Closing this PR as the changes are now in the main branch.