Is your feature request related to a problem? Please describe.
We need to crop the science data before PSF subtraction so we're not wasting computation/memory. This would be best done as a separate step function.
Describe the solution you'd like
This step should crop the data (and error/dq arrays) to the new desired shape and update the pixel locations of the array center, mask location, star location (if computed), etc in the header.
Additional context
We can assume a square image shape if only one integer is passed. Consider if we should assert that the image shape evenness/oddness is preserved to avoid half-pixel shifts for centering.
We can also assume that we are cropping to the outer working angle of the instrument mode (plus a 10% or so border) if no specific shape is passed. We should have a dictionary of the OWA for each mode stored somewhere in the DRP code. The imaging mode can be determined based on the coronagraph focal plane mask name (at least for the modes we're supporting so far).
Is your feature request related to a problem? Please describe. We need to crop the science data before PSF subtraction so we're not wasting computation/memory. This would be best done as a separate step function.
Describe the solution you'd like This step should crop the data (and error/dq arrays) to the new desired shape and update the pixel locations of the array center, mask location, star location (if computed), etc in the header.
Additional context