Add --wait flag to babs status

[asmacdo]

Add --wait flag to babs status

• Depends on babs-status: replace pandas with dataclasses #359

Problem

babs status is a snapshot — it prints the current state and exits.
Automation needs to wait for jobs to finish before running babs merge, so end-to-end users must write wrapper scripts that poll squeue.
Workarounds exist, such as the walkthrough-tests.sh using an squeue loop, but users shouldn't have to implement that themselves, and that approach doesn't update job_status.csv.

Changes

Add --wait and --wait-interval flags to babs status:

# Poll until all submitted jobs complete or fail (default: check every 5 min)
babs status --wait

# Custom interval
babs status --wait --wait-interval 30

Behavior:

• Prints job status summary each iteration (same as babs status)
• Exits 0 when all submitted jobs have results
• Exits 1 if any jobs failed, or if no jobs have been submitted
• Exits 130 on Ctrl-C (clean handling)
• Project-scoped: only watches jobs from this BABS project, not all user jobs

Test changes

• 8 unit tests for babs_status_wait covering success, failure, mixed results, polling loop, no-submitted-jobs, report-each-iteration, and Ctrl-C handling
• E2E walkthrough tests now use babs status --wait instead of manual squeue polling loops
• E2E tests validate job_status.csv after babs merge — every submitted job must have has_results=True and is_failed=False

Housekeeping

• Move pytest-timeout from conda environment files to pyproject.toml (separate commit)

TODO:

• handle SIGINT from con-duct, ctrl+c isnt working
• handle case where not all jobs are submitted, currently exits 1 if not all jobs submitted and finished

Test plan

• Unit tests pass (pytest tests/test_interaction.py)
• E2E walkthrough tests pass in container CI

[codecov-commenter]

Codecov Report

❌ Patch coverage is 93.10345% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 79.73%. Comparing base (912ef8a) to head (cb9f079).

Files with missing lines	Patch %	Lines
babs/cli.py	60.00%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #354      +/-   ##
==========================================
+ Coverage   79.55%   79.73%   +0.18%     
==========================================
  Files          17       17              
  Lines        1956     1984      +28     
  Branches      331      335       +4     
==========================================
+ Hits         1556     1582      +26     
- Misses        279      280       +1     
- Partials      121      122       +1     

☔ 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.

[asmacdo]

I've successfully used this on 2 jobs that passed. A third job failed, and this change caused a problem. I fixed a babs-status bug (babs status didn't update jobs that were stuck in R or PD). My failing job took too long to fail to try again, but I added an integration test in for that case.

If CI is green, I think this one's ready. found another bug :(
fixed by rebase on 359

[asmacdo]

FAILED tests/test_babs_workflow.py::test_babs_init_raw_bids[session] - Failed: Timeout >300.0s

But that one seems flakey, seen it time out a handful of times before.

[asmacdo]

New issue: succeeded jobs now marked as succeeded and failed. fixed by rebase on 359

[asmacdo]

After rebasing on the modified babs status rewrite, I ran this successfully with my demo dataset, I think we are ready!

[asmacdo]

hit 2 smallish issues, updated description with TODO