Add a test collection workflow pre-step

[ThomasLaPiana]

Closes #2033

Code Changes

• add a nox command to collect tests
• add a workflow job that runs the test collection nox session
• make all test-related workflows conditional on the Collect-Tests workflow step via the Build step

Steps to Confirm

• run nox -s collect_tests locally (first run will take awhile as it caches the installations)
• verify the Collect-Tests job runs as expected and the output looks correct
• verify that if the Collect-Tests doesn't pass, the pytest jobs don't get run

Pre-Merge Checklist

• All CI Pipelines Succeeded
• Documentation:
  • documentation complete, PR opened in fidesdocs
  • documentation issue created in fidesdocs
• Issue Requirements are Met
• Relevant Follow-Up Issues Created
• Update CHANGELOG.md

Description Of Changes

This is meant to be a straightforward PR that will serve to both

1. Transparently show code coverage for the entire test suite on every run (at least when analyzed statically)
2. Fail fast if there are any test collection errors and block the expensive pytest jobs from running

[ThomasLaPiana]

the above CI check shows that when the collection passes, everything runs as usual.

I'll now purposefully break the collection to confirm that the tests wont run

[ThomasLaPiana]

the above CI run should break and prevent pytests from running

edit: note that they were skipped as expected

[sanders41]

LGTM!

This is good like it is, just a question. Would it be worth it to move the collection test to the docker build since that step is what takes longest to run and the pytest steps won't run without build?

[ThomasLaPiana]

LGTM!

This is good like it is, just a question. Would it be worth it to move the collection test to the docker build since that step is what takes longest to run and the pytest steps won't run without build?

Oh, that makes way more sense! Ok, i'll make build rely on collect-tests

[sanders41]

The skipped and failed tests here look to me to be github network related and not PR related. Restarting to see if that fixes it.

[sanders41]

The failure again looks unrelated to the PR