Open
Description
I couldn't think of how one would use the CFA that we're given by one of the various raw decoders, but uh, I think I figured it out? It wasn't that hard I was just lost.
It really is just looking around the current pixel for the values we need and then getting them. Hardcoding in their positions will still be faster, I imagine, so maybe we check to see if we have one hardcoded before calling back to the works-for-everything.
Some testing to see speed differences would be fun, too!
(this would also allow us to operate on non-cfa-aligned raw crops, which could be nice)