JP-3734: Changed Default CI Test Algorithm to OLS_C

[kmacdonald-stsci]

Resolves JP-3734

Closes #

This PR addresses default algorithm testing for ramp fitting. The OLS_C algorithm is now the default algorithm for testing.

This PR also addresses a bug that resulted in incorrect attempts to do the CHARGELOSS recalculations when not needed, which would result in a crash when trying to dereference a NULL pointer for rd->orig_gdq.

The PR revealed another bug in the creation of the optional results product. The arrays were being created using PyArray_EMPTY, which didn't initialize memory, so unused elements in some of the optional results product arrays were junk. The creation function used now is PyArray_ZEROS, which initializes the array to zero.

Tasks

• [ ] update or add relevant tests
• [ ] update relevant docstrings and / or docs/ page
• [ ] Does this PR change any API used downstream? (if not, label with no-changelog-entry-needed)
  • [ ] write news fragment(s) in changes/: echo "changed something" > changes/<PR#>.<changetype>.rst (see below for change types)
  • [ ] run regression tests with this branch installed ("git+https://github.com/<fork>/stcal@<branch>")
  • [ ] jwst regression test
  • [ ] romancal regression test

news fragment change types...

- ``changes/.apichange.rst``: change to public API - ``changes/.bugfix.rst``: fixes an issue - ``changes/.general.rst``: infrastructure or miscellaneous change

[codecov[bot]]

Codecov Report

Attention: Patch coverage is 86.95652% with 6 lines in your changes missing coverage. Please review.

Project coverage is 81.80%. Comparing base (5f94030) to head (a83e20b). Report is 10 commits behind head on main.

Files with missing lines	Patch %	Lines
tests/test_ramp_fitting.py	77.77%	6 Missing :warning:

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #298 +/- ## ========================================== - Coverage 85.19% 81.80% -3.40% ========================================== Files 46 46 Lines 8804 8810 +6 ========================================== - Hits 7501 7207 -294 - Misses 1303 1603 +300 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[kmacdonald-stsci]

Regression test:

https://plwishmaster.stsci.edu:8081/job/RT/job/JWST-Developers-Pull-Requests/1769/

Edit: No failures for this test run.

[kmacdonald-stsci]

I recreated the orig_gdq error noted in

https://github.com/spacetelescope/jwst/issues/8842

using the main branch using the command

strun calwebb_detector1 jw01523003001_03102_00001_mirifulong_uncal.fits --steps.ramp_fit.maximum_cores=half

I verified this PR fixes this error.

[tapastro]

@schlafly I believe this PR touches code that Roman also uses, and it looks like Dave is OoO. Are you willing to review this?

[schlafly]

Thanks. We turned off the "ordinary" ramp fitting unit tests in romancal ~last month, and for our regression tests we use the Casertano+22 stuff, so I doubt this affects Roman. But I am running some regtests now.

[schlafly]

This ran fine on the Roman side, thanks! https://github.com/spacetelescope/RegressionTests/actions/runs/11238326482/job/31242982194