Calculate mean source positions using astropy

[mauch]

Currently the mean positions of Sources are derived from a weighted average of the individual source RA and Dec in DD. Although most of the time this is fine due to the small angle approximation for the input - it may be innacurate at the poles and can cause issues with RA wrapping at 0 deg.

The simple fix would be to use astropy to derive the positions instead.

This is a minor issue and not high prority but is perhaps something to fix on a quiet day in future.