Fix slow /api/v1/jobs endpoint by optimizing RunDao queries

[jni-bot]

Summary

Fixes #2987 — the /api/v1/jobs endpoint takes 7+ minutes per job at scale due to an unoptimized query in RunDao.findByLatestJob.

The root cause: the findByLatestJob WHERE clause joins runs_view with jobs_view to find matching run UUIDs. This forces PostgreSQL to build the full runs_view (~2.3M rows) and scan the entire run_facets table (~21M rows) and dataset_facets table (~22GB) in the LEFT JOIN subqueries before filtering — even though only 1-10 runs are needed.

Changes

RunDao.java:

• New findCurrentRunByJob method — uses jobs.current_run_uuid (added in Add col current_run_uuid to jobs #2929 but never leveraged here) for a direct indexed UUID lookup. Returns only the single latest run.
• Optimized findByLatestJob method — replaces the runs_view JOIN jobs_view WHERE clause with a filter on runs.job_uuid which hits the existing composite index runs_job_uuid ON runs(job_uuid, transitioned_at DESC). Includes symlink resolution for aliased jobs.

JobDao.java:

• findAllWithRun (jobs list page) now calls findCurrentRunByJob with limit=1 instead of findByLatestJob with limit=10. This is the hot path — called once per job on every list page load.

Why this works

Path	Before	After
Jobs list (per job)	Full runs_view ⨝ jobs_view scan → 7 min	current_run_uuid indexed lookup → ms
Job detail	Full runs_view ⨝ jobs_view scan → 7 min	job_uuid composite index scan → seconds

Test plan

• RunDaoTest.testFindByLatestJob — passes (validates symlink resolution)
• RunDaoTest full suite — all 6 tests pass
• JobDaoTest full suite — all 8 tests pass
• spotlessApply — clean

[boring-cyborg]

Thanks for opening your first pull request in the Marquez project! Please check out our contributing guidelines (https://github.com/MarquezProject/marquez/blob/main/CONTRIBUTING.md).

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 81.18%. Comparing base (a89b89c) to head (c9062b2).

Additional details and impacted files

@@            Coverage Diff            @@
##               main    #3091   +/-   ##
=========================================
  Coverage     81.18%   81.18%           
  Complexity     1506     1506           
=========================================
  Files           268      268           
  Lines          7356     7356           
  Branches        325      325           
=========================================
  Hits           5972     5972           
  Misses         1226     1226           
  Partials        158      158           

☔ 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.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.