Handling larger range for BUNIT conversions

[e-koch]

This is to address #683. There are a number of fairly common unit transitions that we've been missing. We also have some gaps for different objects that can be addressed here (e.g., 2D Projection or Slice).

SpectralCube

• Jy/beam <-> Jy/pix
• Jy/sr <-> Jy/beam
• K <-> Jy/sr
• K <-> Jy/pix

VaryingResolutionSpectralCube

• K <-> Jy/beam
• Jy/beam <-> Jy/pix
• Jy/sr <-> Jy/beam
• K <-> Jy/sr
• K <-> Jy/pix

Projection/Slice

• K <-> Jy/beam
• Jy/beam <-> Jy/pix
• Jy/sr <-> Jy/beam
• K <-> Jy/sr
• K <-> Jy/pix
• K km/s <-> Jy/beam (km/s) (integrated intensity)
• [ ] km/s <-> m/s (centroid/line width) (Unsure why I added this. It's a basic transform that already works.)

OneDSpectrum

• K <-> Jy/beam
• Other unit conversions require spatial WCS info which we arent keeping in the slice for now (see WIP: Keep full 3D WCS objects for LDOs #395 )

VaryingResolutionOneDSpectrum

• K <-> Jy/beam

[codecov-io]

Codecov Report

❗ No coverage uploaded for pull request base (master@71fd8e2). Click here to learn what that means.
The diff coverage is n/a.

@@            Coverage Diff            @@
##             master     #700   +/-   ##
=========================================
  Coverage          ?   77.66%           
=========================================
  Files             ?       24           
  Lines             ?     5579           
  Branches          ?        0           
=========================================
  Hits              ?     4333           
  Misses            ?     1246           
  Partials          ?        0           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 71fd8e2...4891302. Read the comment docs.

[e-koch]

All brightness unit conversions should be handled and tested now.

Integrated units like K km/s aren't working yet.

[codecov-commenter]

Codecov Report

Merging #700 (1a53cb7) into master (8bb9a37) will decrease coverage by 3.68%.
The diff coverage is 86.36%.

❗ Current head 1a53cb7 differs from pull request most recent head a27f2f4. Consider uploading reports for the commit a27f2f4 to get more accurate results

@@            Coverage Diff             @@
##           master     #700      +/-   ##
==========================================
- Coverage   80.30%   76.62%   -3.69%     
==========================================
  Files          24       24              
  Lines        5551     5475      -76     
==========================================
- Hits         4458     4195     -263     
- Misses       1093     1280     +187     

Impacted Files	Coverage Δ
spectral_cube/cube_utils.py	80.71% <84.93%> (+0.55%)	⬆️
spectral_cube/lower_dimensional_structures.py	81.41% <87.50%> (-2.56%)	⬇️
spectral_cube/base_class.py	85.17% <100.00%> (-2.61%)	⬇️
spectral_cube/spectral_cube.py	75.91% <100.00%> (-8.05%)	⬇️
spectral_cube/visualization-tools.py	0.00% <0.00%> (-12.25%)	⬇️
spectral_cube/ytcube.py	18.81% <0.00%> (-10.90%)	⬇️
spectral_cube/dask_spectral_cube.py	84.60% <0.00%> (-4.59%)	⬇️
spectral_cube/masks.py	79.41% <0.00%> (-4.03%)	⬇️
... and 5 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 8bb9a37...a27f2f4. Read the comment docs.

[e-koch]

@astrofrog @keflavich -- This is ready to review for BUNIT conversions. Composite unit conversions like Jy/beam km/s -> K km/s are not working, but I'm in favour of splitting that into another PR.

[keflavich]

Only nitpicks.

Per our discussion, don't merge until #709 is solved

[keflavich]

WHOA, I didn't know u.one is u.dimensionless! neat!

[e-koch]

Tests passing now that #709 is merged. The remaining failures for the windows build are from a yt issue.

[keflavich]

Can we add at least one specific by-hand correctness check?