Make which quantities are sampled more readily configurable

[dylanhmorris]

This PR creates a clearer interface for configuring what quantities are sampled during fitting and posterior prediction. This will make it easier to use the same infrastructure for the various Pyrenew-{H,E,W} combinations.

To do:

• Unit tests to confirm that configuration of what is sampled works as expected
• Fix end to end test

[codecov]

Codecov Report

Attention: Patch coverage is 66.66667% with 19 lines in your changes missing coverage. Please review.

Project coverage is 27.69%. Comparing base (df353de) to head (b8dc990).

Files with missing lines	Patch %	Lines
pipelines/forecast_state.py	0.00%	11 Missing ⚠️
pipelines/fit_pyrenew_model.py	0.00%	4 Missing ⚠️
pipelines/generate_predictive.py	0.00%	4 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #326      +/-   ##
==========================================
+ Coverage   24.47%   27.69%   +3.22%     
==========================================
  Files          22       24       +2     
  Lines        1704     1755      +51     
==========================================
+ Hits          417      486      +69     
+ Misses       1287     1269      -18     

Flag	Coverage Δ
hewr	41.68% <ø> (ø)
pipelines	13.86% <61.22%> (+9.18%)	⬆️
pyrenew_hew	29.05% <100.00%> (+1.08%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

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

[damonbayer]

What do you think about an approach that parsed the model name? We already use model_name as an argument to many functions in the pipeline.

[damonbayer]

What do you think about an approach that parsed the model name? We already use model_name as an argument to many functions in the pipeline.

Per F2F discussion, the model name approach will not be taken in this PR. In the future, we could implement is and make use of the switches approach implemented in this PR.