Implementation of ePSF and astrometry fix

[gmzsebastian]

This pull request is intended to be for the STIPS 2.0 release, implementing a number of changes. Mainly, the implementation of an ePSF routine that fixes previous astrometry issues. Other changes include:

• Forcing versions of specific packages to be compatible with each other.
• Inclusion of a "bright" and "extra bright" parameters to allow for larger PSFs to be generated for bright sources.
• Estimates for detector offsets for all 18 SCAs.
• Changed detector size to 4088 x 4088.
• Removed filter F149.
• Updated PHOTFNU values.
• Changed how rounding of floats is done.
• Introduction of makePSF.py file with all the calculations related to generating the PSF and ePSF

Closes #155 Closes #142 Closes #141 Closes #134 Closes #130 Closes #129 Closes #128 Closes #111 Closes #110

[york-stsci]

I'm going to take a look at the environment.yml file, and see if I can figure out what's going on, because it seems as if the checks are failing more or less entirely because the environment isn't building for some reason.

[york-stsci]

The CI test failures with python 3.7 are part of removing python 3.7 support from STIPS version 2.0 I will file a cleanup PR that removes outdated tests. As such, and because the testing under python 3.8 succeeds both on GitHub and locally, I approve this PR.

The above is also a waiver for requiring passing CI tests for this particular PR. Note that STIPS will still need to have a clean set of tests for the 2.0.0 release.