fix(jobs): show collect stage progress

[mihow]

Summary

When jobs start, they often transition to STARTED, but show now progress and the stages stay gray / unstarted. This is usually because the "collect" stage is running, but the status was not updated when the collect stage started.

Two small, independent fixes bundled in one PR:

• fix(jobs) — _run_pipeline_job updates the collect stage to STARTED in memory but doesn't save() until after collect_images() returns. For large collections that call can run for minutes, so polling clients (UI) see a frozen job. Add job.save() right after marking the stage STARTED.
• chore(jobs) — commit the deferred 0023_alter_job_job_type_key migration. This is the cosmetic choices-label drift from 585cd55a (collection → capture set rename) that was explicitly excluded from Refactor job logging to use separate table #1259 to keep that diff focused. No schema change; just keeps makemigrations --check green.

Before:

After:

Test plan

• python manage.py makemigrations --check passes
• Trigger an ML job against a large collection; confirm the UI shows the collect stage as in-progress while collect_images() is still running (instead of appearing frozen until collection finishes)

🤖 Generated with Claude Code

Summary by CodeRabbit

• Bug Fixes

  • Job state now persists immediately upon initialization of the collection stage, ensuring status updates are reflected without delay.

• Chores

  • Updated job type field constraints to enforce valid job type identifiers.

[netlify]

✅ Deploy Preview for antenna-preview canceled.

Name	Link
🔨 Latest commit	f1eb099
🔍 Latest deploy log	https://app.netlify.com/projects/antenna-preview/deploys/69f3f292fe7e4900087e1908

[netlify]

✅ Deploy Preview for antenna-ssec canceled.

Name	Link
🔨 Latest commit	f1eb099
🔍 Latest deploy log	https://app.netlify.com/projects/antenna-ssec/deploys/69f3f29214b99e0008865179

[coderabbitai]

📝 Walkthrough

Walkthrough

This PR adds a Django migration to define allowed choices and constraints for the Job.job_type_key field, and modifies the ML job image collection flow to persist the job immediately after initializing the "collect" stage, ensuring earlier visibility of state changes.

Changes

Cohort / File(s)	Summary
Job Type Field Migration ami/jobs/migrations/0023_alter_job_job_type_key.py	Defines choices (ml, populate_captures_collection, data_storage_sync, unknown, data_export, post_processing), constraints (max_length=255), default value ("unknown"), and verbose name ("Job Type") for the Job.job_type_key field.
Job Persistence Logic ami/jobs/models.py	Persists job to database immediately after initializing the "collect" stage to STARTED with progress=0, ensuring state changes are visible earlier in the ML job image collection flow.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Possibly related PRs

• fix(jobs): finalize async job when 0 images need processing #1191: Both PRs modify ami/jobs/models.py to handle job state persistence when image processing occurs, ensuring jobs are persisted/flushed at appropriate points in the flow.

Poem

🐰 This rabbit hops with code so true,
Job types constrained to just a few,
Now save each stage without delay,
State changes visible right away! 🎯

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Title check	✅ Passed	The title 'fix(jobs): show collect stage progress' directly describes the main fix in the changeset, which persists job state immediately after marking collect stage as STARTED to show progress in the UI.
Description check	✅ Passed	The pull request description follows the template structure with Summary, List of Changes, Related Issues, Detailed Description, How to Test, Screenshots, and Deployment Notes sections. All required sections are present and substantively completed.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch fix/job-progress-collect-stage-save

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[Copilot]

Pull request overview

This PR improves job progress visibility during long-running ML “collect” operations and adds a deferred migration to keep job_type_key choices in sync after the “collection” → “capture set” rename.

Changes:

• Persist collect stage STARTED progress immediately so polling clients see active progress during collect_images().
• Add migration 0023_alter_job_job_type_key to update the job_type_key choice label (“Populate captures collection” → “Populate capture set”) and keep makemigrations --check clean.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
ami/jobs/models.py	Saves the job immediately after marking the collect stage as STARTED, preventing UI polling from appearing frozen during long collection runs.
ami/jobs/migrations/0023_alter_job_job_type_key.py	Alters job_type_key field choices to reflect updated human-readable label for the populate job type.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.