add spectral unit to moment map 0 display unit, simplify logic

[cshanahan1]

There was previously an issue where the moment map plugin was using the spectral y axis unit for moment 0 rather than always using a surface brightness unit. This was fixed in #3156.

This PR applies some of the changes from #3116, which also fixed that issue but based on an outdated main pre many unit conversion changes, so I found it easier to cherry pick some of the code here rather than rebasing that PR.

The display unit now has the spectral unit because specutils is pinned to the version with the fix for MM units. I made some small tweaks including using _get_display_unit rather than grabbing the UC plugin from the MM plugin and accessing display unit there.

I think sufficient test coverage was added in #3156 so i didn't add the tests from #3116, but i did expand one of the tests to test that the moment map plugin has the correct units when spectral unit is changed (previously was just testing when flux unit was changed)

This will close JDAT 4651 and #3116

Closes #3116.

[codecov]

Codecov Report

Attention: Patch coverage is 94.11765% with 1 line in your changes missing coverage. Please review.

Project coverage is 88.50%. Comparing base (21e9b87) to head (135a58b).
Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
...configs/cubeviz/plugins/moment_maps/moment_maps.py	93.75%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3211      +/-   ##
==========================================
- Coverage   88.51%   88.50%   -0.01%     
==========================================
  Files         125      125              
  Lines       18694    18697       +3     
==========================================
+ Hits        16547    16548       +1     
- Misses       2147     2149       +2     

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

[kecnry]

is this (and the other removal below) because the marks are handling the unit conversion themselves?

[cshanahan1]

yes i believe so (but also cc @bmorris3) . I also verified this change doesn't break anything in line analysis where this mixin is also used.

[cshanahan1]

also see https://jira.stsci.edu/browse/JDAT-4830, this bug was not introduced with this change

[kecnry]

see #3212 for the fix to moment map continuum previews during unit conversion. Whichever PR makes it in second should rebase and ensure that is still working as expected.

[kecnry]

LGTM - would like to have @bmorris3 review before merging, if possible, just to make sure we're not missing anything from that original PR.

[bmorris3]

Here are some differences with #3116 for @cshanahan1 to double check. Otherwise all good by me.

[bmorris3]

Some changes not carried over from #3116 for review:

1. https://github.com/spacetelescope/jdaviz/pull/3116/files#diff-44f616436de8315fb25244c8555ea83596e03bedf23685221f4becc69bad8fbbR32
2. https://github.com/spacetelescope/jdaviz/pull/3116/files#diff-44f616436de8315fb25244c8555ea83596e03bedf23685221f4becc69bad8fbbR112
3. (not sure if this one is still needed?) https://github.com/spacetelescope/jdaviz/pull/3116/files#diff-44f616436de8315fb25244c8555ea83596e03bedf23685221f4becc69bad8fbbR121-R122

[cshanahan1]

1. fixed, removed Flux from options
2. i see this correctly removing flux as an option already
3. I intentionally omitted this because I found it to work without it, and I new @kecnry did some work recently on units with an empty viewer so I figured thats why it was now working

[cshanahan1]

I have no idea why 3.12 tests are failing

[pllim]

I think scipy dev wheel is broken again. You can ignore it.

[cshanahan1]

in that case im going to merge this, if i can find some consequence of not subscribing to AddDataMessage i'll follow up but i don't see any issues