Handling cosmic rays

[sfr348]

Two classes of problems with new CR handling:

A: Are real CR properly handled? This involves ensuring the following:

1. CR filtering is sufficiently aggressive
2. Number of frames is sufficient to make up SNR in masked pixels from CR filtering

B: Are non-CR features incorrectly masked? This involves:

1. If CR filtering is 'on' and CR rate is 0, what happens?
2. Do we need to update corgihowfsc to use the entire FS area when calculating EETC parameters to ensure centre bright speckle is not always saturated and mislabeled as a CR? (see image below)