This fixes the combined bounding box location for regions bigger than 256x256.
It also fixes the position on taking coalesced objects in the non-root regions, where position checks are properly done on rez instead.
It also fixes the megaregion land channel to return null if the land does not exist, which should probably also be done for the ordinary land channels rather than returning a dummy region.
Inspiration from Garmin's commit in http://opensimulator.org/mantis/view.php?id=6595. Thanks!
However, the calls to the land management module to record prims need to remain, since they were also being used to return owner object lists, etc.
This is probably why prim counts were being done there in the first place.
The region spanning parcel shouldn't exist in this situation. If it does, when the land data is loaded it is repersisted with a local ID which comes after the ones loaded via the oar, which obliterates the oar loaded one.
Resaving the data we just loaded from the database is somewhat odd in itself (though this makes sense from the way that OAR loading was already using the same event).
