Bring the docs up to speed

[trontrytel]

This PR brings the docs compile time back to reasonable. 112.712372 seconds (1.54 G allocations: 72.581 GiB, 5.14% gc time, 36.17% compilation time: 8% of which was recompilation) on my laptop. The reason for the slowdown was setting a very demanding tolerances for the ODE solver in our parcel model (especially if running with Float64 precision). I think by default we should run everything in Float32

Btw, when I was running individual simulations I noticed some of them were throwing warnings like:

WARNING: Multiple deposition nucleation parameterizations chosen to run in one simulation. Parcel will only run MohlerAF_Deposition for now.

Is that expected behavior @amylu00 ?

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (dfcdce7) 98.66% compared to head (999f005) 98.66%.
Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #324   +/-   ##
=======================================
  Coverage   98.66%   98.66%           
=======================================
  Files          33       33           
  Lines        1047     1047           
=======================================
  Hits         1033     1033           
  Misses         14       14           

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

[trontrytel]

It takes about 11 minutes to get all the dependencies and less than 4 minutes to generate the docs. We could make it shorter by for example deciding on using only one package for plotting. And by making the solver tolerances in parcel even less restrictive.

But maybe it's good enough. When building things locally it should be faster (about 2 minutes for me, not counting instantiating)

[amylu00]

Oh that might be wrong. I'll try fixing it sometime, it shouldn't be showing up for the examples we currently have. I think I mistakenly used or conditional statements when it should've been &&. Since there's so many modes being added, it'll have to be done another way anyways though. Thanks for the heads up!

[amylu00]

Looks good to me

[trontrytel]

Oh that might be wrong. I'll try fixing it sometime, it shouldn't be showing up for the examples we currently have. I think I mistakenly used or conditional statements when it should've been &&. Since there's so many modes being added, it'll have to be done another way anyways though. Thanks for the heads up!

Sounds good. I'll open an issue so that we don't forget