[ENH] Add Level objects to channels.json for motion

[effigies]

• Replaces [ENH] Propose to use *_channels.json for the description of coordinate systems #1524.

This was initially comments on #1524, but it became a lot and easier to just make a proposal. Making pull requests against other people's branches often fails, so might as well just make an alternative PR.

The things worth noting:

1. This is essentially taking allow Levels values (in sidecar for TSV files) to be objects #1573 as accepted. I'm fine with it, but this is a change from the current spec, where these fields MUST be strings. This probably needs a PR on its own.
   
2. I think adding "n/a" is strictly worse than omitting a field. In one case I can test for presence.
3. "ReferenceFrameDescription" is redundant when it is in sidecar.reference_frame.Levels.<rf_name> object. I just changed it to Description.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (28beccd) 87.83% compared to head (6ab3e00) 87.83%.
Report is 82 commits behind head on master.

❗ Current head 6ab3e00 differs from pull request most recent head 4645fd1. Consider uploading reports for the commit 4645fd1 to get more accurate results

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1591   +/-   ##
=======================================
  Coverage   87.83%   87.83%           
=======================================
  Files          16       16           
  Lines        1356     1356           
=======================================
  Hits         1191     1191           
  Misses        165      165           

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

[sjeung]

Thank you, I agree with all of the proposed changes.

Example data sets in this PR reflects them already (renamed ReferenceFrameDescription -> Description).
Once Julius approves my change to spec this PR will be up to date with it too. (Removed unnecessary n/a options and renamed ReferenceFrameDescription -> Description)

1. This is essentially taking allow Levels values (in sidecar for TSV files) to be objects #1573 as accepted. I'm fine with it, but this is a change from the current spec, where these fields MUST be strings. This probably needs a PR on its own.

Do you plan to open this one or should I have a look into implementing this?

[effigies]

Do you plan to open this one or should I have a look into implementing this?

That was done in #1603.

[sappelhoff]

I think mentioning Description is a good idea, but feel free to ignore if you disagree. Anyhow, +1 to merge