Getting DREAMS working!

[hugobuddel]

This PR should make it possible to simulate things with DREAMS.

The main two changes are:

• A workaround / hack to make it possible to simulate the full 6 square degrees of DREAMS. ScopeSim currently only works up to 1 square degree. A proper fix is implemented in Enlarge initial Field of View to suppord wide-field imagers AstarVienna/ScopeSim#433 . The workaround can therefore be removed once that is merged and a new ScopeSim version is released.
• An increase of the FWHM of the PSF from 1.5 arcsec to 15 arcsec. Apparently ScopeSim does not work well when the PSF is about the size of a pixel (2.48 arcsec). I'm not particularly happy with this solution though.

Some minor changes:

• The file J.dat is moved to the filters directory, as that is where DREAMS.yaml claims it should be.
• A figure with all 6 detectors is added.
• An empty ExtraFitsKeywords effect is added. This prevents the default keywords from being added, which was raising quite some errors.
• The star field is enlarged to fill the entire field of view, otherwise the outer detectors do not show anything.

About the PSF size: I'm not really sure how to deal with the small PSF size. I've now arbitrarily made it 10 times larger. I have not investigated the cause of the problems with the small PSF. It could even be that ScopeSim is working fine, but that matplotlib downsamples the image. I don't know. How important is the correct PSF for you @Astrogirlanajli ?

Can you move forward with this for now?

[hugobuddel]

Forgot to add a figure:

[hugobuddel]

ScopeSim does actually work with the smaller FWHM. You just can't see it in the matplotlib figures because the stars are all only a pixel or so.

Manually setting the Low and High cutlevels in skycat does show the sources:

Crop:

I've also turned on cmds["SIM.sub_pixel.flag"] = True. That should work slightly better with the small PSF. But I believe it is only 4 sub-pixels per pixel, so it's not perfect.

I've added a fits file (dreams.detector2.fits) of one of the detectors to this branch. Since it is possible to see the stars, I declare this a success!

What are your thoughts @Astrogirlanajli ?