Fix suffix list test for new nsclean step

[hbushouse]

This PR fixes a failing unit test, due to the recent addition of the new "nsclean" step.

Checklist for maintainers

• added entry in CHANGES.rst within the relevant release section
• updated or added relevant tests
• updated relevant documentation
• added relevant milestone
• added relevant label(s)
• ran regression tests, post a link to the Jenkins job below.
  How to run regression tests on a PR
• Make sure the JIRA ticket is resolved properly

[hbushouse]

Latest commit also fixes codestyle failures and updates the README for the B10.0.1 installation in Ops yesterday.

[codecov]

Codecov Report

Attention: 2 lines in your changes are missing coverage. Please review.

Comparison is base (f77b5d6) 75.42% compared to head (989f2d6) 75.42%.
Report is 1 commits behind head on master.

Files	Patch %	Lines
jwst/nsclean/lib.py	0.00%	2 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #8111   +/-   ##
=======================================
  Coverage   75.42%   75.42%           
=======================================
  Files         464      464           
  Lines       37938    37938           
=======================================
  Hits        28615    28615           
  Misses       9323     9323           

Flag	Coverage Δ		*Carryforward flag
nightly	77.37% <ø> (ø)		Carriedforward from f77b5d6

*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.

[tapastro]

Do we know why the nsclean PR didn't catch this?

LGTM

[hbushouse]

Do we know why the nsclean PR didn't catch this?

LGTM

Which problem: the unit test or the codestyle? The codestyle error showed up in the original PR and I just ignored it. Don't recall seeing the unit test failure.

[tapastro]

Do we know why the nsclean PR didn't catch this?
LGTM

Which problem: the unit test or the codestyle? The codestyle error showed up in the original PR and I just ignored it. Don't recall seeing the unit test failure.

I meant the unit test - I went back and looked at the PR and the unit tests passed then. Seems strange.

[hbushouse]

Do we know why the nsclean PR didn't catch this?
LGTM

Which problem: the unit test or the codestyle? The codestyle error showed up in the original PR and I just ignored it. Don't recall seeing the unit test failure.

I meant the unit test - I went back and looked at the PR and the unit tests passed then. Seems strange.

It's one of those little mysteries of life.

[braingram]

Looks like I was too slow :)

LGTM. Thanks for the request for a review as a good excuse to look at the suffix code. It looks pretty straightforward to add a new one.

[hbushouse]

LGTM. Thanks for the request for a review as a good excuse to look at the suffix code. It looks pretty straightforward to add a new one.

Well, it may look that way on the surface, but it's always taken me a couple of tries to get it right.

[jdavies-st]

It would be really good to make the suffix replacement code more flexible. Either a registry, or dynamic based on registered steps in the stpipe entry_points. Currently, if I add a new step, register it via entry points, and specify a class_alias and suffix, it's suffixes are still not replaced properly, and there's no way as a user to fix this except for explicit string replacement.

Also, having _nsclean and _nscleanstep be default suffixes for the same step, and the path through stpipe being the only thing that triggers one vs the other, is not great. Better for most steps to have a defined, default suffix. And in fact, for that defined, default stuffix to be the same as class_alias for the step or pipeline. This won't be possible for all steps and pipelines due to historic usage, but it will be possible be for most, and certainly for new steps.

Also, it would be really good to standardize that the step_defs dict in pipelines have keys that are the same as class_alias for each step, for consistency.

I'll file a separate issue on this.