Investigate and implement image distortion model for Galileo SSI

[rfrenchseti]

Currently the GOSSI SSI host instrument does implement any image distortion model. The report Klaasen (1993) "Galileo Solid-State Imaging Subsystem Calibration" shows an error of around 1.19 pixels at the corner. Given all of the other problems with SSI, it's not clear at the moment if that's really important to fix, but we should consider it.

[rfrenchseti]

My experiments show a twist of -0.067 +/- 0.024 degrees, which is about 0.6 pixels in the corner. A distortion model might fix this. There doesn't appear to be major distortion otherwise, but there are a very small number of star fields to examine, and they are of generally poor quality.