Smooth and prune logs in nightly process

Made-for-purpose IoT databases normally prune measurements after a day or so (or even after a week, after a month) to remove measurements that aren't useful, like temperatures that didn't change between two measurements and locations that didn't change or represent measurement noise. We can remove the extra locations around the hab in this process.

Remove gps points from display when preceded and followed by close neighbors

Outliers look like a spike out and back, so this is a good experiment to try to remove outliers. It probably needs to be done before the geofencing that could eliminate the points we need to examine. That suggests the geofencing is done in python after this process is finished.

Start by locating the nightly process and the Python geofencing stage described in the issue. Trace how GPS points reach display and determine where consecutive close neighbors and spike-shaped outliers are handled. Done means the intended extra locations and noise points are removed without eliminating points needed for geofencing.

Written by the indexing model from the issue text.