make SkyModel inherit from UVBase, adjust accordingly

[bhazelton]

Description

Make SkyModel inherit from UVBase. This highlighted some inconsistencies with how we were handling different spectral types. This PR clarifies that and makes a number of changes to fully support all the different spectral types.

Motivation and Context

closes #37
closes #24
closes #20
closes #12

partially addresses #8 -- Flux still needs units.

Checklist:

• I have read the contribution guide.
• My code follows the code style of this project.
• Any new or updated docstrings use the numpy docstring format.
• I have added tests to cover any changes.
• My change includes a breaking change
  • My change includes backwards compatibility and deprecation warnings (if possible).
• I have updated the CHANGELOG if appropriate.

[codecov]

Codecov Report

Merging #49 into master will increase coverage by 3.78%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master      #49      +/-   ##
==========================================
+ Coverage   91.40%   95.19%   +3.78%     
==========================================
  Files           4        4              
  Lines         512      853     +341     
==========================================
+ Hits          468      812     +344     
+ Misses         44       41       -3     

Impacted Files	Coverage Δ
pyradiosky/utils.py	100.00% <ø> (ø)
pyradiosky/skymodel.py	97.61% <100.00%> (+2.98%)	⬆️

---

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 73693e6...26d58f5. Read the comment docs.

[bhazelton]

@aelanman Ok, I think this PR is finally ready for you to look at.

[aelanman]

This looks great! Thank you for all of this work. My comments are mostly questions and a few nitpicks. I think the biggest thing we'll have to figure out following this is how to handle the votable catalog reading in pyuvsim again.

I haven't looked very closely at the changes to test_skymodel.py, but from what I can tell they look fine. I'd like to try running the reference simulations with these changes to see if they break anything before approving. We may need to include a small patch to pyuvsim just to get the gleam catalog to work, before doing a larger overhaul of how catalogs are specified in configuration files.

[mkolopanis]

I took a general scroll through and things seem to make sense to me, nothing glaring stuck out. I can give it a much finer toothed comb after lunch today.

[mkolopanis]

I was able to look more closely. Changes Look nice.

[aelanman]

Thanks for addressing my comments. I just have a couple more minor things I noticed, but this should be ready to go.

[aelanman]

This looks good! Thanks for doing all this.

[aelanman]

@bhazelton I've changed test_circumpolar_nonrising to actually run source_cuts and check that the rise/set lsts, and that sources that will never rise are excluded by the cut and sources that are circumpolar are identified correctly.

[aelanman]

This all looks good to me now, but Bryna should take a look at the test I changed.

[bhazelton]

@aelanman your test looks good. I added a round-tripping test for the rise/set_lst parameters through the conversion to a SkyModel object and back. I also added support (and a test) for MoonLocation type telescope locations.

[aelanman]

Looks good! Thanks.