Issue 537: Fix making analysis dataframes and record priors

[SamuelBrand1]

This draft PR is for tracking changes to the EpiAwarePipeline code base done in order to run the full pipeline and to fix unanticipated changes in the analysis code.

I've decided to upgrade the work so far to a full PR. This PR closes #537 and contributes to #405.

Tasks so far

• Record the tightened priors chosen to partially address Model specific priors capability #405; but still general to all models.
• Fix make_prediction_dataframe_from_output function to implement changes to saved objects due to Upgrade the testmode pipeline script to pipeline scripts: prior predictive + full #445. This closes Issue 536: Fix make dataframe function #537

[github-actions]

Try this Pull Request!

Open Julia and type:

import Pkg
Pkg.activate(temp=true)
Pkg.add(url="https://github.com/CDCgov/Rt-without-renewal", rev="latest-analysis", subdir="EpiAware")
using EpiAware

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 90.54%. Comparing base (5274760) to head (a9e823e).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #535   +/-   ##
=======================================
  Coverage   90.54%   90.54%           
=======================================
  Files          57       57           
  Lines         825      825           
=======================================
  Hits          747      747           
  Misses         78       78           

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

[SamuelBrand1]

This PR modifies the make_prediction_dataframe_from_output function to:

1. Fix the function interface to the output data structure by responding to pipeline changes.
2. Use new inference_config rather than logic on the filename of serialised object.

This PR also adds a small "test data" in the correct format for use in unit testing. From f2f discussion with @seabbs this is a temporary solution until we make a test which generates its own test data.

[SamuelBrand1]

I was thinking about breaking this up, but I didn't like the sequence and ended up with a with set of commits.