Fix numba JIT disable patch targeting wrong config attribute

[Tomaz-Vieira]

Test suite tried to force-disable numba JIT when already imported by patching NUMBA_DISABLE_JIT on numba.core.config, but numba internally reads the env var into an attribute named DISABLE_JIT (no prefix).

We should find a less hacky solution to this, but for now, running the tests sequentially (pytest -x) always pass.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 92.41%. Comparing base (09b0bcb) to head (099eb3b).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1181   +/-   ##
=======================================
  Coverage   92.41%   92.41%           
=======================================
  Files          52       52           
  Lines        7874     7874           
=======================================
  Hits         7277     7277           
  Misses        597      597           

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Tomaz-Vieira]

pre-commit is flagging a mypy error that has nothing to do with this PR, so I don't think we should pollute it with a fix

[LucaMarconato]

Thanks for the fix, well spotted!

pre-commit is flagging a mypy error that has nothing to do with this PR, so I don't think we should pollute it with a fix

Yes, clean solution is to fix it in another PRs, but practically those pre-commit fail happen routinely so we usually just address them at the first occasion (or sometimes merge to main with red CI and fix it later, if we know that we are going to fix it soon anyway). I fixed it here right before merging: #1180.