Simplify checks for 1D/2D spectra and traces

[rosteen]

Now checks for the importer in these cases like a lot of the other is_valid checks, and additionally adds a _data_type field in meta to distinguish cases where a collapsed 1D spectrum is being loaded by the 2D or 3D spectrum importer. This is purely internal so I don't think it needs a changelog, and it should be covered by existing tests.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 84.74%. Comparing base (02ddc71) to head (e2a1e8e).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #4176      +/-   ##
==========================================
- Coverage   84.76%   84.74%   -0.03%     
==========================================
  Files         205      205              
  Lines       30552    30554       +2     
==========================================
- Hits        25897    25892       -5     
- Misses       4655     4662       +7     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[MatthewPortman]

This is, on first glance, confusing... any objection to moving the data_type instantiations into __init__ via self.data_type = ... for all of these importers?

[rosteen]

You know, it made more sense when I was going to check data_type for all of the validity checks instead of sticking with looking at _importer in most cases. I think I can simplify this for now since the only times I'm looking at _data_type are in the override cases, and not set it on the importer class at all. Unless it's something we do want to start adding to meta for every importer.