Swap to new model

[jvshields]

📝 Description

Type: 🚀 feature | ☣️ breaking change | 🎢 infrastructure

This PR attaches swaps the old model created by read_marcs_to_fv() out and hooks up the new model reader to the run_stardis() function. This effectively implements much of the restructure work with the new infrastructure, and so it also includes changing the code in many places to read information from the new model structure.

Future PRs need to change the Configuration schema to allow for general model reading, as the current code only allows for unzipped MARCS models, so the configuration should allow to specify whether the input model is gzipped or not.

Additionally, I have left many comments around the code pointing out parts that need cleaning or changing, because they are misleading or redundant. These are not-functional changes, and should be changed in a future cleaning PR.

I am leaving the read_marcs_to_fv() function for now, because it is currently integrated into the benchmarking pipeline, but a future PR should also swap that over to the new structure as well.

🚦 Testing

How did you test these changes?
I have tested the output of the quickstart notebook with the same model as before. I get a different answer for the spectrum, shown here

There is a very small difference here that is not entirely unexpected. The previous model reader made physical changes to the system by naively interpolating the temperatures and densities of the system and then defining cells with those values, and solving an opacity for those cells. This new method solves the plasma using the values provided by the model, and then linearly interpolates the opacities measured at each of the points for light to travel from one depth point to the next. I generally believe that the new model is working correctly, but as far as I can tell it's impossible to directly reproduce the previous model because that model solves a completely different plasma and then solves the radiative transfer equation differently.

The answers are close enough that I think the new model is working correctly.

• Testing pipeline
• Other method (describe)

☑️ Checklist

• I requested two reviewers for this pull request
• I updated the documentation according to my changes
• I built the documentation by applying the build_docs label

Note: If you are not allowed to perform any of these actions, ping (@) a contributor.

[codecov]

Codecov Report

Merging #112 (0282128) into main (a26fe04) will decrease coverage by 0.16%.
Report is 2 commits behind head on main.
The diff coverage is 2.50%.

❗ Current head 0282128 differs from pull request most recent head cce9b27. Consider uploading reports for the commit cce9b27 to get more accurate results

@@            Coverage Diff             @@
##             main     #112      +/-   ##
==========================================
- Coverage   34.04%   33.89%   -0.16%     
==========================================
  Files          17       17              
  Lines         658      658              
==========================================
- Hits          224      223       -1     
- Misses        434      435       +1     

Files Changed	Coverage Δ
stardis/base.py	33.33% <0.00%> (-4.17%)	⬇️
stardis/io/model/marcs.py	67.70% <0.00%> (-0.72%)	⬇️
stardis/model/geometry/radial1d.py	66.66% <0.00%> (ø)
stardis/opacities/base.py	13.71% <0.00%> (+0.23%)	⬆️
stardis/opacities/broadening.py	31.08% <0.00%> (ø)
stardis/plasma/base.py	44.61% <ø> (+0.67%)	⬆️
stardis/transport/base.py	25.00% <0.00%> (-0.50%)	⬇️
stardis/model/base.py	11.36% <50.00%> (ø)

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[sonachitchyan]

Looks good! The only thing is maybe we should move the comments to an issue and make a separate PR for solving them? This way it feels disconnected from the purpose of this PR