Allowing custom gain value, sinusoid_pc with PyMC3 models in S5, and bug fixes

[taylorbell57]

This PR:

• Allows users to provide their own gain value to Stage 3 (useful for MIRI where the value on CRDS is/was very wrong)
• Allows the use of the sinusoid_pc model with the theano-based differentiable versions fit with PyMC3. This allows for faster fitting of phase variations compared to the starry spherical harmonics as it ignores eclipse mapping signals. This can be useful when you expect no eclipse mapping signals or want to compare fits with and without accounting for eclipse mapping signals to ensure they're significant
• Updates the image_registration package to 0.2.6 (now that it's on PyPI)
• Fixes an issue with determining the maximum order for spherical harmonics that caused a crash when using odd-m Ylm terms and also excessively slowed the code when there were only even-m Ylm terms
• Fixes issues with the differentiable models related to shared fits
• Forces the eccentricity to be between [0,1] for PyMC3 models (otherwise the code crashes)
• Fixes an issue with the batman eclipse model that's only present when there is a non-zero eccentricity
• Fixes a bug with saving S5 outputs in the savetable_S5 function
• Fixes the calculation of the offsets and amplitudes in Stage 6 for PyMC3 models

There are no currently open issues that this PR closes.

[codecov-commenter]

Codecov Report

Merging #505 (54b479b) into main (847822c) will decrease coverage by 0.99%. The diff coverage is 26.79%.

:mega: This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

@@            Coverage Diff             @@
##             main     #505      +/-   ##
==========================================
- Coverage   59.57%   58.59%   -0.99%     
==========================================
  Files          87       90       +3     
  Lines       10188    10465     +277     
==========================================
+ Hits         6070     6132      +62     
- Misses       4118     4333     +215     

Impacted Files	Coverage Δ
...rve_fitting/differentiable_models/CentroidModel.py	34.09% <0.00%> (ø)
...reka/S5_lightcurve_fitting/models/CentroidModel.py	28.57% <0.00%> (ø)
src/eureka/S6_planet_spectra/s6_spectra.py	49.53% <0.00%> (-0.40%)	:arrow_down:
src/eureka/S1_detector_processing/bias_sub.py	12.65% <12.65%> (ø)
src/eureka/S1_detector_processing/s1_process.py	13.72% <16.66%> (+0.18%)	:arrow_up:
...itting/differentiable_models/SinusoidPhaseCurve.py	19.37% <19.37%> (ø)
src/eureka/lib/astropytable.py	71.42% <27.27%> (-15.67%)	:arrow_down:
src/eureka/S5_lightcurve_fitting/s5_fit.py	76.40% <33.33%> (-0.85%)	:arrow_down:
src/eureka/S1_detector_processing/superbias.py	46.15% <46.15%> (ø)
...curve_fitting/differentiable_models/StarryModel.py	79.45% <50.00%> (-0.14%)	:arrow_down:
... and 4 more

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