December card reset update

[bwgref]

This takes care of (some of) the card reset issues.

First, the clock drift 250 ms off during the first December issue.

This looks like this:

We've "fixed' this by flagging all of the out-of-family points as "bad". Then we also had to insert a break point at MET 408299422 in the spline.

The most recent data are still a little worse than I'd like. But I think that will get better with time. Right now, things have not gotten back to "good" since the card resets.

[bwgref]

Forgot to add stuff to app.py. Did that, and added a second break point. It now looks like this, which I'm pretty happy with.