Sp reimplement

[jcapriot]

Summary

Reimplements the SP method in SimPEG.

PR Checklist

• If this is a work in progress PR, set as a Draft PR
• Linted my code according to the style guides.
• Added tests to verify changes to the code.
• Added necessary documentation to any new functions/classes following the
  expect style.
• Added relevant method tags (i.e. GRAV, EM, etc.)
• Marked as ready for review (ff this is was a draft PR), and converted
  to a Pull Request
• Tagged @simpeg/simpeg-developers when ready for review.

What does this implement/fix?

This reimplements the SP code in SimPEG, making heavy use of the existing DC formulation. It is written that the q property is the model.

It also defines a few new maps that are specific to the SP methodology, one relating the secondary current density vector defined at cell-centers to q by taking the negative divergence of it. The second relates the hydraulic head to the q. These two maps are simple extensions of the LinearMap. Together with the simple formulation relating to q these should mimic all of the previous functionality of the SP code.

Additional information

I went back and forth deciding on the name of the module this lives, i.e. it is either spontaneous_potential or self_potential. I went in favor of spontaneous purely because the word self has a common meaning in python programming. But I'm happy either way.

[jcapriot]

@sgkang Here is my implementation of the SP simulation. Very lightweight as it inherits almost everything from the DC Simulation.

[jcapriot]

I'm not sure this has a place in SimPEG right now...

[sgkang]

no need to bring this in.

[codecov]

Codecov Report

Merging #1211 (ffbf663) into main (b3b2ca9) will decrease coverage by 0.19%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main    #1211      +/-   ##
==========================================
- Coverage   81.99%   81.81%   -0.19%     
==========================================
  Files         159      162       +3     
  Lines       24411    24481      +70     
==========================================
+ Hits        20016    20028      +12     
- Misses       4395     4453      +58     

Impacted Files	Coverage Δ
.../electromagnetics/static/resistivity/simulation.py	96.49% <100.00%> (+0.01%)	⬆️
...magnetics/static/spontaneous_potential/__init__.py	100.00% <100.00%> (ø)
...gnetics/static/spontaneous_potential/simulation.py	100.00% <100.00%> (ø)
...omagnetics/static/spontaneous_potential/sources.py	100.00% <100.00%> (ø)

... and 3 files with indirect coverage changes

[sgkang]

Looking great. Your SP implementation is concise and elegant.
Played a little bit with the sparse inversion, and it seems working well.

It seemed the sensitivity (or Ainv) is not updated while inversion, but it would be nice if you could check.

It would be great to add some tutorials at least for each of three source types, and further replicate Craig's field example.

Above example is accessible through: https://curvenote.com/@sgkang09/researches/sp-3d-inv-charge.2

[jcapriot]

Thanks Seogi for the review. I've made the requested updates and will pull this in. Feels good to not have the leftover pieces from before the simulation switchover in the codebase anymore!

We can add tutorials and such in another PR.