JP-2669: Fixing the NaN and DO_NOT_USE bug.

[kmacdonald-stsci]

Updating handling NaNs in the rateints product, as well as the DQ flags associated with those NaNs.

Resolves JP-2669

Closes #

This PR addresses NaNs in the rateints product file and the DQ flags associated with them. In the rateints product, NaNs were present with Good DQ flags. They shouldn't be.

Further, INS requested no NaNs be present in the rateints file. This PR detects NaNs in the rateints file and sets them to 0.0, then sets the corresponding DQ flag to DO_NOT_USE.

Lastly, the rateints DQ flags were set considering only JUMP_DET and SATURATED. This has now been changed to consider DO_NOT_USE as well. Specifically, if all groups in an integration for a pixel are flagged as DO_NOT_USE, the corresponding DQ value will be flagged as DO_NOT_USE.

Checklist

• added entry in CHANGES.rst (either in Bug Fixes or Changes to API)
• updated relevant tests
• updated relevant documentation
• updated relevant milestone(s)
• added relevant label(s)

[codecov]

Codecov Report

Merging #112 (10fcc1c) into main (5013a69) will decrease coverage by 0.02%.
The diff coverage is 66.66%.

@@            Coverage Diff             @@
##             main     #112      +/-   ##
==========================================
- Coverage   72.13%   72.11%   -0.03%     
==========================================
  Files          16       16              
  Lines        2494     2510      +16     
==========================================
+ Hits         1799     1810      +11     
- Misses        695      700       +5     

Impacted Files	Coverage Δ
src/stcal/ramp_fitting/ramp_fit_class.py	55.05% <14.28%> (-2.78%)	⬇️
src/stcal/ramp_fitting/ols_fit.py	72.06% <100.00%> (+0.08%)	⬆️
src/stcal/ramp_fitting/utils.py	84.10% <100.00%> (+0.27%)	⬆️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[dmggh]

LGTM, including your documentation updates.

[hbushouse]

Be aware that there's related discussion going on in https://jira.stsci.edu/browse/JP-2692 as to what values to assign to pixels that have no usable data (i.e. NaN vs. zero vs. ???). I believe this is scheduled for Cal WG discussion next week, so this might need updating based on the outcome of that discussion.

[karllark]

Looks good to me. Very minor comment.

[hbushouse]

Merging this in order to fix the bug that was leading to incorrect results. If the Cal WG decides later that pixels with no good data should be set to something other than zero, we'll make that change when necessary.

[nden]

This was merged without an approval by the Roman team.
@ddavis-stsci fyi: I started the roman devdeps regtest job.

[hbushouse]

This was merged without an approval by the Roman team. @ddavis-stsci fyi: I started the roman devdeps regtest job.

Whoops, my bad!

[kmacdonald-stsci]

I expect this to cause failures for regtests, since it changes NaNs to 0.0 and sets corresponding DQ flag to DO_NOT_USE.