Drizzlepac/HAP: Confirm alternate solution contains distortion correction

[stscijgbot-hstdp]

Issue HLA-1603 was created on JIRA by Steve Goldman:

In STWCS PR #⁠266 we have addressed issues related to a change in astropy PR #⁠19228. 

With both of these changes, we still need to make sure that in all cases a distortion correction is still applied. 

It may be worth adding a regression test to insure that this is always the case. 

[stscijgbot-hstdp]

Comment by Steve Goldman on JIRA:

I've updated the test_apply_as_alternate_method test in STWCS PR #⁠246 to actually change the alternate WCS before comparing the old and new versions, making the test more robust. 

Most modern WFC3 and ACS images use lookup tables for the distortion correction. Since this corresponds to header keywords CPDIS2  = 'Lookup ", in most cases the secondary distortion correction will look the same as the primary.

Where this may cause an issue is when applying a secondary distortion correction using hlet.apply_as_alternate on much older data (e.g. WFPC2 images) that use a geometric distortion calibration file (DGEOFILE) that differs. Since this is a very rare an unlikely case, we will ignore it for the time being.