Normalize patches by the star-center value rather than the patch maximum

[svank]

Btw if you're gearing up for another release, there was one last local change I never pushed. When the cutouts of each star are being collected, each one gets normalized. The was done by dividing each patch by its maximum value. In cases where there's some contaminant in the patch that's brighter than the star, the normalized star is in the range [0, small] instead of [0, 1] and so doesn't contribute well to the final PSF model. I changed it to use the patch-center value (i.e. the star's center and presumably peak value) rather than the patch's maximum value, to help ensure that the stars are all normalized the same. Any bright contaminants will be normalized to values > 1, but since they're hopefully rare, that shouldn't really affect the median/percentile/whatever.

If the patch sizes are small, hopefully these contaminants are pretty rare and this is a low-impact change.

What do you think?

[codecov]

Codecov Report

Patch coverage: 100.00% and no project coverage change.

Comparison is base (93ab305) 94.57% compared to head (4d58b0b) 94.57%.

Additional details and impacted files

@@           Coverage Diff           @@
##             main      #60   +/-   ##
=======================================
  Coverage   94.57%   94.57%           
=======================================
  Files          11       11           
  Lines         996      996           
=======================================
  Hits          942      942           
  Misses         54       54           

Files Changed	Coverage Δ
regularizepsf/fitter.py	90.00% <100.00%> (ø)
tests/test_fitter.py	100.00% <100.00%> (ø)

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

[jmbhughes]

This is a pretty minor so I probably can just include it. It seems like a good idea. It speeds things up (likely very minor since the patches are small) by also not computing a max of a patch but just accessing one value. I'll think about any unforeseen side effects and I'll merge early next week probably.

I changed it to merge into develop. Once we have all the improvements assembled, we'll merge into main and roll a release.