Add Total CI Duration benchmark to anneal workflow

[joshlf]

Motivation

• Capture and publish the total CI runtime for the Anneal Tests job in addition to existing pull and test timing metrics.
• Provide a single metric that measures end-to-end job duration to help track regressions in overall CI execution time.

Description

• Add a Record job start time step that writes a Unix timestamp to GITHUB_OUTPUT as unix.
• Update the Combine benchmarks step to read steps.job_start_time.outputs.unix into JOB_START_TIME_UNIX, compute the total CI duration, emit total_time.json, and combine pull_time.json, test_time.json, and total_time.json into output.json using jq.
• Preserve existing benchmarking upload step benchmark-action/github-action-benchmark to store the new Total CI Duration (All Steps) metric alongside the other CI duration metrics.

Testing

• Executed the updated GitHub Actions workflow for the Anneal Tests job in CI and verified output.json includes the new Total CI Duration (All Steps) entry.
• Confirmed the benchmarking upload step completed successfully and the job finished without errors.

---

Codex Task

[github-actions]

🤖 Optimistically Approved: Changes appear scoped. Final strict verification will occur in the Merge Queue.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 91.88%. Comparing base (9329ada) to head (0497a2d).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #3323   +/-   ##
=======================================
  Coverage   91.88%   91.88%           
=======================================
  Files          20       20           
  Lines        6076     6076           
=======================================
  Hits         5583     5583           
  Misses        493      493           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.