Implement capping of catalog MAXMAG to 11.2 mag

[taldcroft]

Based on discussions in emails ("maxmag" near June 7, 2019), the MAXMAG parameter in commanded star catalogs should be capped at 11.2 mag. This makes that change in proseco catalogs.

This PR does not implement capping of the maxmag parameter found in the acq star selection code. That value of maxmag is actually a different quantity, defined as 1-sigma above the star mag. In practice that will rarely be above 11.2 for selected stars, but in any case I would rather leave the acq selection algorithm unchanged so this PR does not introduce any change in star selection.

cc: @mbaski @jskrist

[jeanconn]

Should this just be cut as proseco 4.5? I don't think there are any doc or other action items to close out with regard to the new maxmag.

[taldcroft]

I'm going to defer to your judgement as Ska release manager. I'm honestly not sure about the MATLAB release schedule, if this should go in to the Matlab release, or when, and etc. The handling of this is not entirely clear cut as you probably realized.

[jeanconn]

OK. I think we should just cut this as a release, but I didn't consider the maxmag change as a bugfix. This would mean a 4.5 release; I think we haven't really decided if that means we need to do anything with 4.3.x and 4.4.x branches.

[jeanconn]

Oh and I suppose we need to update the ACA load review checklist ACA-041 and also update the starcheck check to match .