[CLI][Sugestion] Adding flags for evals to return only failed and print to output file

[jottakka]

Adding to arcade evals:
-f or --failed-only: shows only failed evals
-o or --output: write to an output txt file

---

Note

Adds failed-only filtering and file output to arcade evals, refactors result display to support console/file with accurate summaries, and adds tests; bumps version to 1.6.0.

• CLI evals:
  • Add flags --failed-only,-f and --output,-o.
  • Filter results via utils.filter_failed_evaluations, passing original counts to display.
• Display:
  • Extract _display_results_to_console; support failed_only disclaimer and original-count summary.
  • Add output_file writing (creates parent dirs) with same formatted output as console.
• Utils:
  • New filter_failed_evaluations(all_evaluations) returning filtered data and (total, passed, failed, warned).
• Tests:
  • Add libs/tests/cli/test_display.py and libs/tests/cli/test_main_evals.py covering display details, failed-only mode, file output, and filtering.
• Version:
  • Bump project version to 1.6.0.

^{Written by Cursor Bugbot for commit baad441. This will update automatically on new commits. Configure here.}

[codecov]

Codecov Report

❌ Patch coverage is 93.75000% with 4 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
libs/arcade-cli/arcade_cli/main.py	0.00%	4 Missing ⚠️

📢 Thoughts on this report? Let us know!

[jottakka]

Moved changes to here: #689