STIPS 2.2 Pull Request

[gmzsebastian]

This pull requests includes a number of bug fixed and improvements to STIPS for its 2.2 release, including:

• Updated the requirements to Pandeia 3.1
• Updated the Pandeia reference data.
• Fixed two bugs with extended sources, one with the orientation of the WCS, and another that was causing all galaxies to be generated with sersic index n = 1.
• Included a feature where users can add a boolean flag "fast_galaxy" to use an older, faster, less accurate, method for injecting extending sources.
• Updated the documentation to reflect the new "fast_galaxy" capability.

[ojustino]

Hello @gmzsebastian, thank you for your work in this PR. I've addressed most of the findings of my code review in gmzsebastian/STScI-STIPS#6, so please check it out. Other than that, I have a few questions:

• Could you please update CHANGES.rst here with entries for 2.2.0? These should include the upgraded minimum versions of Pandeia (and Python, if you accept gmzsebastian/STScI-STIPS#6).
• Do we still need to specify that webbpsf==1.1.0 exactly or is it possible to work with any of the newer versions? RSP would probably appreciate the flexibility in the latter case, but I don't think it should delay the v2.2.0 release if not.
• What do you think of switching the repository's primary feature branch from master to main to bring it in line with other repositories in the ST/astropy universe? I don't think that action is technically linked to the release, so I can do it after the release is over if you agree.

[ojustino]

Thanks for merging my pull request. What do you think about the questions in my last comment here? (And about the possibility of listing a number for how much less accurate the fast_galaxy scenario is in the examples.rst?) I believe we're ready for a merge once those have been addressed.