Better diangostics for parallel test execution in CI

[potiuk]

When running tests in parallel in our CI we switched off showing success outputs by default in #38157. This is generally a good thing, because it only shows outputs for the test types that failed and makes the CI log output far snappier in case only one or few test types failed.

However it also has a side effect that it's
not really easy to see what kind of commands were run in each parallel test type, so we might get accidentally some mistakes where we miss that our tests are not doing what we think they are doing - for example not running tests for various Python versions, only for one, or not running them for various database versions (both cases happened in the past). By hiding the success outputs, we make it harder to spot such issues.

This PR introduces two ways to improve it:

• we can now set a label on PR to include success outputs and any committer can set the label to show success outputs as well.

• When we run parallel unit tests we summarize what is going to be executed - showing all the ShellParams passed to the parallel test execution. This is the best place to show it because this ShellParams object contains all the actual values passed - after processing throught options, special environment variables and so-on - so those are the actual values used to run the parallel tests.

Also it will be printed in a prominent place - just before test execution in the main console output, which means it will be visible without the need to unfold any output.

---

^ Add meaningful description above
Read the Pull Request Guidelines for more information.
In case of fundamental code changes, an Airflow Improvement Proposal (AIP) is needed.
In case of a new dependency, check compliance with the ASF 3rd Party License Policy.
In case of backwards incompatible changes please leave a note in a newsfragment file, named {pr_number}.significant.rst or {issue_number}.significant.rst, in newsfragments.

[potiuk]

That's how it will look like: