JP-3121: Updating JWST Ramp Fit Code to Work with New C Extension

[kmacdonald-stsci]

Resolves JP-3121

This PR addresses the C extensions. In particular, tests need to ensure the dtypes of ndarrays are what is to be expected in a RampModel. Tests have been changed to ensure proper typing, as well as simplified.

Checklist for maintainers

• added entry in CHANGES.rst within the relevant release section
• updated or added relevant tests
• updated relevant documentation
• added relevant milestone
• added relevant label(s)
• ran regression tests, post a link to the Jenkins job below.
  How to run regression tests on a PR
• Make sure the JIRA ticket is resolved properly

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 7 lines in your changes are missing coverage. Please review.

Project coverage is 57.97%. Comparing base (781e0e0) to head (81f6271).
Report is 9 commits behind head on master.

Files	Patch %	Lines
jwst/ramp_fitting/ramp_fit_step.py	0.00%	7 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #8355      +/-   ##
==========================================
+ Coverage   57.93%   57.97%   +0.04%     
==========================================
  Files         387      387              
  Lines       38839    38831       -8     
==========================================
+ Hits        22502    22514      +12     
+ Misses      16337    16317      -20     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[hbushouse]

Started regtest run at https://plwishmaster.stsci.edu:8081/job/RT/job/JWST-Developers-Pull-Requests/1461

[hbushouse]

@kmacdonald-stsci Regtest results at https://plwishmaster.stsci.edu:8081/job/RT/job/JWST-Developers-Pull-Requests/1461/testReport/ show differences in just one NIRISS test, in the rate and rateints files. The VAR_RNOISE array shows a smallish number of pixels with different values. Can you confirm whether this is expected? This was run using the latest stcal/main, which includes the new C code for ramp_fit, but with it deactivated (so it should've run the existing python version).

[kmacdonald-stsci]

@kmacdonald-stsci Regtest results at https://plwishmaster.stsci.edu:8081/job/RT/job/JWST-Developers-Pull-Requests/1461/testReport/ show differences in just one NIRISS test, in the rate and rateints files. The VAR_RNOISE array shows a smallish number of pixels with different values. Can you confirm whether this is expected? This was run using the latest stcal/main, which includes the new C code for ramp_fit, but with it deactivated (so it should've run the existing python version).

Using the master branch in JWST and main STCAL results in those tests passing, so the failures aren't due to STCAL. The failures are due to the changes in the ramp_fit_step.py in this PR. In the master branch, the entire read noise array is updated due to CHARGELOSS recomputation of read noise variances, whereas in this branch only the pixels flagged as CHARGELOSS are updated. The differences in the regression tests are expected.

[hbushouse]

Code updates and regtest results look good.