Dommaschk potentials implemented in JAX

[dpanici]

Following Dommaschk 1986 paper, implementing general Dommaschk potentials in JAX so can automatically differentiate to find the field

TODO:

• Add tests for each function added
• Add new MagneticField subclass DommaschkMagneticField which uses autodiff to grad(dommaschk_potential) to find the B field
• replace CD_mk and CN_mk with the expressions in eqns 31 and 32 of the paper, so they are applicable for any m,k
• Possibly verify against simsopt implementation - I don't think this is needed, I also verified thru field tracing that we reproduce Fig. 2 in the 1986 dommaschk paper.
• jaxify for loops in dommaschk potential functions
• other ways to speed up evaluation or fitting?
• implement stellarator symmetry option for fitting ( makes $a_{ml}=d_{ml}=0$ for even $l$ and $b_{m,l}=c_{m,l}=0$ for odd $l$, so effectively reduced number of modes by 2, useful for fitting)
• put magnetic_fields in its own directory and break out dommaschk utility functions?

Pushing off improving the fit logic to #928 , this PR has been good to go for a while now, might as well get it in.

[codecov]

Codecov Report

Attention: Patch coverage is 99.71182% with 1 lines in your changes are missing coverage. Please review.

Project coverage is 94.71%. Comparing base (ba62bb2) to head (5830208).

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #483      +/-   ##
==========================================
- Coverage   94.96%   94.71%   -0.25%     
==========================================
  Files          83       86       +3     
  Lines       21012    21198     +186     
==========================================
+ Hits        19954    20078     +124     
- Misses       1058     1120      +62     

Files	Coverage Δ
desc/compute/_surface.py	94.49% <ø> (ø)
desc/compute/data_index.py	95.55% <ø> (ø)
desc/magnetic_fields/__init__.py	100.00% <100.00%> (ø)
desc/magnetic_fields/_core.py	84.32% <100.00%> (ø)
desc/magnetic_fields/_dommaschk.py	100.00% <100.00%> (ø)
desc/objectives/_free_boundary.py	96.80% <ø> (ø)
desc/magnetic_fields/_current_potential.py	99.39% <99.39%> (ø)

[ddudt]

Approving, but check docstring formatting