RCAL-776: Change image units.

[PaulHuwe]

Resolves RCAL-776

Closes #1114 #1123

This PR changed image units in code and tests. Added gain reduction.

NOTE: no regression test link, because these changes require update to regression test files.

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

All modified and coverable lines are covered by tests ✅

Project coverage is 75.74%. Comparing base (bd465ec) to head (18c2f03).
Report is 1 commits behind head on main.

❗ Current head 18c2f03 differs from pull request most recent head 0b394f9. Consider uploading reports for the commit 0b394f9 to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1128      +/-   ##
==========================================
- Coverage   76.72%   75.74%   -0.99%     
==========================================
  Files         105      105              
  Lines        7054     7045       -9     
==========================================
- Hits         5412     5336      -76     
- Misses       1642     1709      +67     

Flag	Coverage Δ		*Carryforward flag
nightly	62.70% <ø> (-0.04%)	⬇️	Carriedforward from 6c8a97f

*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]

This should wait until the rad and roman_datamodels changes are in. We don't really want the pyproject to point to your local branches.
The code changes look fine.

[PaulHuwe]

This should wait until the rad and roman_datamodels changes are in. We don't really want the pyproject to point to your local branches. The code changes look fine.

Yeah, they are temporary just for tests to pass. I have updated the Changelog.

[schlafly]

This case isn't the important one since we don't use this mode, but instead of adding this code, I think it would be better to delete lines 118-122 and 128-140.

[PaulHuwe]

I was afraid to remove those in case it affected calculations being made.

[schlafly]

No, the earlier lines do the exact opposite of what you're doing here? We plan to get rid of the support for the old ramp fitting mode eventually anyway, but I don't see why we should multiply in the gain and then divide it back out without doing any work in the meantime? Do you see something I'm missing?

[PaulHuwe]

Removed.

[schlafly]

Staring at the schema, there is also var_flat, which we don't actually populate or do anything with. I guess that's a separate ticket.

[schlafly]

This is actually fine and I was somehow missing the flat field uncertainty step computation later on. Disregard this, sorry!

[schlafly]

Looks good to me, thanks.

[ddavis-stsci]

LGTM