TST: Remove custom CI skip

[pllim]

See https://github.blog/changelog/2021-02-08-github-actions-skip-pull-request-and-push-workflows-with-skip-ci/

[codecov]

Codecov Report

Merging #5734 (6e59fbe) into master (bf24eae) will increase coverage by 0.00%.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master    #5734   +/-   ##
=======================================
  Coverage   74.27%   74.27%           
=======================================
  Files         408      408           
  Lines       37129    37134    +5     
  Branches     4571     4571           
=======================================
+ Hits        27577    27583    +6     
+ Misses       9552     9551    -1     

Flag	Coverage Δ		*Carryforward flag
nightly	75.67% <ø> (ø)		Carriedforward from bf24eae
unit	53.16% <ø> (+0.33%)	⬆️

*This pull request uses carry forward flags. Click here to find out more.

Impacted Files	Coverage Δ
jwst/associations/main.py	95.90% <0.00%> (+0.01%)	⬆️
jwst/associations/lib/dms_base.py	92.69% <0.00%> (+0.06%)	⬆️
jwst/tso_photometry/tso_photometry_step.py	77.41% <0.00%> (+1.61%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update bf24eae...6e59fbe. Read the comment docs.

[jdavies-st]

Whoa! They listened to the users!

Now if they would just implement allowed_failure. 😅

[pllim]

Bad news is that currently the built-in CI skipping skips everything in .github/workflow but that isn't a problem for your current use case.

[jdavies-st]

Ah, so if you have a commit at HEAD that has [ci skip], then it will skip the merged job on master plus any cron jobs?

[jdavies-st]

Btw, is there a reason to keep the cancel_ci job in this workflow? Or can all of that be removed now?

[pllim]

Actually, I just merged the equivalent PR on astropy and it looks like the merge commit only has PR title, not the individual commit messages, so the merge ran the CI. I suspect cron would behave the same (I'll find out tomorrow). But if you have [ci skip] in the PR title, you deserve it (LoL).

Well, I have a fix in for cancelling at styfle/cancel-workflow-action#55 but it is just sitting there. You can choose to remove this altogether, keep it as is until my PR is merged (if ever), or use my forked branch; but I think that is out of scope here.

[jdavies-st]

I see. Cool. We'll just leave it here then.

I suspect you would run into the problem if you did "Rebase and merge" which would put that rebased commit with [ci skip] in the title at the HEAD of master. So I see where this could be a problem. We had the same problem back when we had our own Jenkins per-PR CI. Nothing changes.

[jdavies-st]

Looks good. Thanks!

[pllim]

Yeah... at this point, the rule is server side now and out of our control. The only thing we can do is workaround it.