RAD-172 Sky Background

[PaulHuwe]

Resolves RAD-172

Closes #247

This PR adds a sky background schema.

Checklist

• Schema changes discussed at RAD Review Board meeting
• Added entry in CHANGES.rst under the corresponding subsection
• Updated relevant roman_datamodels utilities and tests
• Passed romancal regression testing on Jenkins / PLWishMaster. Link: https://github.com/spacetelescope/RegressionTests/actions/runs/10241364423

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 95.45%. Comparing base (c1811ab) to head (5590671).
Report is 147 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #432      +/-   ##
==========================================
+ Coverage   95.38%   95.45%   +0.06%     
==========================================
  Files           4        4              
  Lines         195      198       +3     
==========================================
+ Hits          186      189       +3     
  Misses          9        9              

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

[ddavis-stsci]

LGTM

[mairanteodoro]

Looks good to me! Thanks, @PaulHuwe! I only have one question about the units for the background level.

[PaulHuwe]

Regression tests have completely passed.

[schlafly]

Thanks. @mairanteodoro , I do want to understand if the DN / s is needed for now or not; it's conceptually wrong but we can defer correction until later if needed.

[mairanteodoro]

@mairanteodoro , I do want to understand if the DN / s is needed for now or not; it's conceptually wrong but we can defer correction until later if needed.

The DN / s unit was just an example (not a requirement) that I provided to @PaulHuwe so that he could start his work; it is absolutely not necessary, since the sky match step will work fine with either unit.

The example that I gave to @PaulHuwe was from a regression test that checks for the results from the SkyMatchStep only. It uses the _cal files from this association file from Artifactory. The data array in those _cal files have all DN / s units and that's what's passed on to SkyMatchStep.

However, the mosaic pipeline runs FluxStep before SkyMatchStep, which fixes the unit issue (everything after FluxStep is in MJy/sr).

[mairanteodoro]

Thanks, @PaulHuwe!