When extracting failure messages "ignore_errors" isn't respected

[d-honeybadger]

What happened?

Had an run that failed due to a timeout (I think?). The error message in this case should be just an exit code and maybe something about context deadline exceeded. But provider-ansible went on to extract potential error messages from the playbook logs (as it should), and found a failure that actually wasn't the reason for the run to fail (below), cause that particular task had ignore_errors: true and therefore the playbook proceeded past this failure.

    message: 'exit status 254: Failed on play "Determine where we are at in the process",
      task "Grab the server CA, if exists", host "test-master-0":
      non-zero return code;"

How can we reproduce it?

Create a run with an ignore_errors: true task but such that the run fails for some other reason (like a timeout).

What environment did it happen in?

Crossplane version:

[d-honeybadger]

Will submit a PR in a bit, just creating this ticket as a TODO for myself