Garbage Commit to Trigger CI (see #2663)

[JacksonBurns]

Title

[alongd]

This one fails after >199 min (one of them at 212, but maybe it's a glitch of the 200 min rule)
In the previous PR we discussed, we know that the regression tests did run originally, and only lately we get this strange behavior.
I'm also at lost, can suspect that one of the dependencies has changes versions, but have no lead.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 54.79%. Comparing base (d426379) to head (576e86b).

❗ Current head 576e86b differs from pull request most recent head bffb858

Please upload reports for the commit bffb858 to get more accurate results.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2667   +/-   ##
=======================================
  Coverage   54.79%   54.79%           
=======================================
  Files         125      125           
  Lines       37156    37156           
=======================================
  Hits        20361    20361           
  Misses      16795    16795           

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

[rwest]

Looks like you're making progress! If you need to commit to main to get the reference artifact uploaded, feel free IMHO.
I noted there's also a warning to update conda-incubator/setup-miniconda@v2 - could be worth doing while you're at it.

[JacksonBurns]

@rwest thanks for the note, have done that here. Interestingly, changing conda-incubator to v3 allowed the macos-latest runner to resolve to macos-14, which runs on the new apple hardware instead of the intel chips. This then caused the CI to fail because none of the RMG channel conda packages are available for the apple hardware.

As a fix now I have set the runner to macos-13, which runs on intel. In the future will would need conda binaries on our channel that are built for apple hardware.

Rather than merge this PR I will just get it to fail the way I want it to (it should discover the correct run of the CI), and then I will move these changes to a branch on my fork, open a new PR, and merge that if it 'fails right'. Just want to make sure this works from forks.

[rwest]

Great.

Yes, let's aim for Intel Mac and Mac Silicon binaries for all conda dependencies - maybe this summer? (Not this PR)

[JacksonBurns]

Yes for sure. #2641 would be a big part of that, since it should allow us to rebuild the RMG binaries for the first time in a while.

[JacksonBurns]

Closing in favor of #2669

[JacksonBurns]

Great.

Yes, let's aim for Intel Mac and Mac Silicon binaries for all conda dependencies - maybe this summer? (Not this PR)

I have started this here: ReactionMechanismGenerator/conda-recipes#18