Improve Dark simulations
We need some raw darks for testing the masterdark recipe.
In particular:
- Make it easier to run them; now it is done through a notebook. Perhaps create a separate script.
- Make them 9 detectors. (Not sure how hard this will be.)
- Make them full size. Also to test scalabity with openmp.
- Add the PRO.CATG in the headers.
- Make it easier to set the mean dark current value; perhaps propagate these in the headers? (They should be in the raw headers.)
- Add the ERR and DQ layer.
- Also create a perfect masterdark and masterflat.
Once the simulator allows it, we should also give them some realistic structure. See #175 (closed)