Careless Accepts SIGI=0

[kmdalton]

Careless actually runs and produces (pathological) output when there are SIGI values of 0 in the input. There should be a way to deal with this by either

1. refusing to run
2. purging entries with SIGI<= 0
3. setting SIGI = maximum(SIGI, np.abs(I) / max_isigi)

I have a slight preference for (3), because i think in general we should discount reported I/SIGI values that are greater than about 100 or so.

[PrinceWalnut]

(3) I'm guessing you meant np.abs(I) / max_isigi? I think this might be worth adding a flag for allowing that in cases of known low-intensity data, but I would personally prefer (2) with a warning label in the logs.

[kmdalton]

yeah! sorry i meant / not *. i updated it.