JP-2500 Allow cube build to accept a tangent point, position angle and number of spaxels in x, y cube dimensions

[jemorrison]

Resolves JP-2500

Closes #6697

This PR addresses allows the user to specify the tangent point (ra_center, dec_center), position angle (cube_pa) and size of cube in x cube dimension (nspax_xi) and y cube dimension (nspax_eta).
These are all option to be run offline not part of standard processing.

Checklist for maintainers

• added entry in CHANGES.rst within the relevant release section
• updated or added relevant tests
• updated relevant documentation
• added relevant milestone
• added relevant label(s)
• ran regression tests, post a link to the Jenkins job below.
  How to run regression tests on a PR
  Results of regression test:
  https://plwishmaster.stsci.edu:8081/job/RT/job/JWST-Developers-Pull-Requests/955/
• Make sure the JIRA ticket is resolved properly

[jemorrison]

@drlaw1558 Test this out and let me know what you think.
I am assuming a user provided ra_center or dec_center is just a smallish change to the current values. If a user gives a very large value (say at the edge of the cube) that becomes the tangent point. Cube build will try and create a cube with the center of the cube at this location. So there can be empty space where no data exist. This can be made smaller by sending in an nspax_xi or nspax_eta but if the tangent point at the edge there still will be empty space. But I am also assuming that most people will not do that and the ra,dec centers are moved just a little bit.
If this PR looks corrrect I will update the docs and add the new terms

[codecov]

Codecov Report

Patch coverage is 44.82% of modified lines.

Files Changed	Coverage
jwst/cube_build/cube_build_step.py	20.00%
jwst/cube_build/ifu_cube.py	57.89%

📢 Thoughts on this report? Let us know!.

[drlaw1558]

LGTM; I tested each of these parameters and combinations thereof, and the results looked as expected. Needs a change log entry though.

[jemorrison]

I will update the documentation now. I just wanted to make sure the code looked correct before I did that.

[hbushouse]

@jemorrison Please run regression tests and post a link to the run here.

[jemorrison]

@drlaw1558 @hbushouse
Howard mentions in a comment about the use of xi and eta. I always think of the tangent plane projection in those coordinates but I realize we never have mentioned xi, eta in the documentation. Should I add a line or two describing that xi, eta are the tangent plane coordinates the cube is build on in the Terminology section of the description or should I change nspax_xi to nspax_x and nspax_eta to nspax_y and refer to those values as the x,y values in the tangent project ? And remove other references to xi and eta ?

[drlaw1558]

@jemorrison Hm, #7783 used the convention scalexy, so by analogy nspax_x and nspax_y would probably make sense. xi/eta and x/y aren't the same values so we need to keep both, but should be interchangeable from the standpoint of numbers of spaxels. That way we wouldn't have to describe xi/eta in anything other than inline documentation as this is just used internally as an intermediate step in the tangent plane projection.

[hbushouse]

Looks good now.

[hbushouse]

I started a regtest run here: https://plwishmaster.stsci.edu:8081/job/RT/job/JWST-Developers-Pull-Requests/932
Theoretically, since this update only involves params used by a user in off-line processing, it should have no effect at all on the regtests. Although perhaps someone should add a new test that utilizes these params.

[jemorrison]

I can add a regression test that sets the size of the cube. Cube build requires a full image to run so I could not figure out a way to test that in a unit test. But I can add a regression tests. I am hesitant to do that right now because of all the regression test failure to run.

[jemorrison]

Results of regression test:
https://plwishmaster.stsci.edu:8081/job/RT/job/JWST-Developers-Pull-Requests/955/

@hbushouse is it ok to merge this - or should I wait since Build 10 is being made now ?