Implement outlier detection step.

[mairanteodoro]

Resolves RCAL-501

This PR addresses the implementation of the outlier detection step for Roman (replaces PR#979).

The link to the regression test results is here.

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: 243 lines in your changes are missing coverage. Please review.

Comparison is base (da2d16a) 73.32% compared to head (15f8520) 71.08%.

Files	Patch %	Lines
romancal/outlier_detection/outlier_detection.py	18.07%	136 Missing ⚠️
...mancal/outlier_detection/outlier_detection_step.py	13.04%	60 Missing ⚠️
romancal/datamodels/container.py	9.52%	38 Missing ⚠️
romancal/resample/resample.py	43.75%	9 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #981      +/-   ##
==========================================
- Coverage   73.32%   71.08%   -2.25%     
==========================================
  Files         103      104       +1     
  Lines        6733     7006     +273     
==========================================
+ Hits         4937     4980      +43     
- Misses       1796     2026     +230     

Flag	Coverage Δ		*Carryforward flag
nightly	63.29% <ø> (ø)		Carriedforward from 7640d00

*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.

[schlafly]

Some minor comments. The most important one: help me with the provenance of the outlierpars reference file, and what the value of pars['resample_data'] is for our case?

[schlafly]

These changes look good. The only outstanding comments are about the outlier_pars file and the value of the resample_data argument. I don't really know how that works or what the defaults are; can you explain what we're doing there? Thanks!

[nden]

I did a superficial review and left inline comments. I still need to run the code and look at the output.

[nden]

The step completes without crashing. I am not sure the results are entirely correct but I think we can merge this as is and debug when we have good data.
The test I ran uses as input 3 identical images (copies of the same file). It found quite a lot of outliers which I think is unexpected.

One thing that needs to change now is adding the OUTLIER DQ flag to this page:
https://github.com/spacetelescope/romancal/blob/main/docs/roman/references_general/dq_flags.inc

[schlafly]

I added the missing flag to the docs, merging.

[mairanteodoro]

Thanks @schlafly and @nden!