running-workflow-name does not work consistently

[egor-yolabs]

From my testing over a couple of weeks, the functionality behind the "running-workflow-name" param is unreliable and will not always wait until all the other checks finish despite what the docs say. It does work most of the time, but there is a race condition. From my observation there is some period of time after a job finishes and another job (with a dependency on the first job) gets queued. If the "wait-on-check-action" just so happens to check the jobs status in that period of time, it will incorrectly conclude that all jobs have finished even though they haven't.

[ramseydsilva]

In my case, just using:

ref: ${{ github.ref }}
check-name: 'Run tests'

also leads to inconsistent results. Sometimes the check incorrectly assumes that the tasks have run. I have noticed that this happens when there are multiple jobs running, but possibly also happens otherwise.