RCAL-800: Set flux step status for each input

[stscieisenhamer]

Resolves RCAL-800

Related PRs:

• roman_datamodel#332
• rad#395

This PR sets the step completion status for each input model to FluxStep.

Checklist

• added entry in CHANGES.rst under the corresponding subsection
• updated relevant tests
• updated relevant documentation
• updated relevant milestone(s)
• added relevant label(s)
• ran regression tests, post a link to the Jenkins job below. How to run regression tests on a PR

[codecov]

Codecov Report

Attention: Patch coverage is 44.44444% with 5 lines in your changes are missing coverage. Please review.

Project coverage is 76.73%. Comparing base (902c5f8) to head (5f2a0bb).
Report is 1 commits behind head on main.

Files	Patch %	Lines
romancal/flux/flux_step.py	44.44%	5 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1160      +/-   ##
==========================================
- Coverage   76.76%   76.73%   -0.03%     
==========================================
  Files         107      107              
  Lines        7096     7101       +5     
==========================================
+ Hits         5447     5449       +2     
- Misses       1649     1652       +3     

Flag	Coverage Δ		*Carryforward flag
nightly	62.80% <ø> (ø)		Carriedforward from 902c5f8

*This pull request uses carry forward flags. Click here to find out more.

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

[ddavis-stsci]

LGTM

[schlafly]

This looks fine to me. Do you think there are cases where we want the pipeline to continue when we have failed to flux calibrate? My sense is that if we try to flux calibrate and fail something has gone badly wrong and we don't want to just continue and mark as skipped, but maybe I'm missing some case where that would be the desired behavior?

[stscieisenhamer]

Sorry for the late reply. I agree that the exception catching may be a bit broad. My intent was to only continue on for the condition of units not being able to convert.

How about if the unit checking it tightened, so that it only checks if the data is already in flux units. Otherwise, if the data is not in the expected DN/s units, fail completely.

[schlafly]

That sounds like a good approach to me.

[schlafly]

This looks good to me. Presumably it will break the regression tests, though, and we'll need to do an okify; are you able to do that?

[stscieisenhamer]

In principle I can and will attempt it.

[stscieisenhamer]

Regtest #707

[schlafly]

Looks good, yup, go ahead and update and okify.