ci: cancel runs if newer commit available and update checkout action

[rem1776]

Description
This adds the same setting Lauren added to the FV3 parallelworks CI. It'll cancel a run in progress if a newer commit is available.

This should help with storage usage and we should have quicker checks with PR's as well since it won't be wasting time on the older commits.

also updates all our checkout actions versions to the newest since the old one was deprecated.

How Has This Been Tested?
ci

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• Any dependent changes have been merged and published in downstream modules
• New check tests, if applicable, are included
• make distcheck passes

[thomas-robinson]

How tested is this? It's not going to cancel a workflow from a different PR, right?

[laurenchilutti]

How tested is this? It's not going to cancel a workflow from a different PR, right?

My understanding is that it will cancel any in process action that matched the conceurrency group. So we name the concurrency group ${{ github.workflow }}-${{ github.ref }} which translates to <name-of-workflow>-<refs/pr/##> for a PR. So this will only cancel in process runs of the same workflow and PR.

I have not tested this on workflow triggered by"push" as in the dycore, we only trigger on PRs

[laurenchilutti]

When I did this for the dycore, I tested it in my PR that brought in the workflow yaml changes (the equivalent of this PR for FMS). You can keep making dummy commits to test that it properly cancels the previous commits CI on this PR

[rem1776]

When I did this for the dycore, I tested it in my PR that brought in the workflow yaml changes (the equivalent of this PR for FMS). You can keep making dummy commits to test that it properly cancels the previous commits CI on this PR

Thanks @laurenchilutti , I'll try that now and check the how it works with the pushes on my fork.

[rem1776]

@laurenchilutti @thomas-robinson looks like its working as intended, will only cancel jobs on the same branch with pushes as well.