Upgrades to Petrosian and Petrosian Corrections [v0.5.0 Core]

[robelgeda]

closes #179, closes #44, closes #109, closes #162, closes #160, closes #123, closes #58, closes #112
Major upgrades to Petrosian workflow. It generalizes the Petrsoian profiles to use any eta and epsilon values.

This PR introduces:

• A new option to address low signal-to-noise ratio issues when measuring r_99 by allowing the use of r_80. While r_80 provides good estimates of r_50 (r_e), it can sometimes result in an overestimation of r_99. To ensure accuracy, correction grids can be utilized to estimate r_99 from r_80. However, we maintain the Petrosian measurement philosophy that correction grids should be reserved for high-precision measurements. Therefore more analytical and testing work is needed before this PR can be merged into production.
• Error estimates for all Petrosian Radii according to Strauss et. al 2002 Appendix A.3.
• Saving/reading profiles to/from file.
• Grid generation now has multi-processing support.

More work is needed to:

• Make correction to 1D radial profiles. This may be out of the scope of this PR.

Other Changes

• Remove kernel key word from segmentation functions since PhotUtils no longer accepts them.
• order_cat moved to segmentation sub-module.
• source_photometry moved to segmentation sub-module.
• make_radius_list moved to utils sub-module.
• Refactor: photometry.flux_to_abmag -> utils.hst_flux_to_abmag
• Refactor: Move measure_fwhm from utils to modeling.fitting
• Depreciate calculate_photometic_density
• Depreciate object_photometry, now its called radial_photometry
• Depreciate measure_fwhm

TODO:

• Improve plots.

Cleanup checks:

• Re-enable doc tests
• Deal with ValueError: A value in x_new is above the interpolation range.