Align Percentile and Exceedance Prob calcs

In trying to write up some documentation, ended up getting into the weeds a bit on the percentile and exceedance probability calculations. Realized right now we are using the Weibull method for percentiles, but not for exceedance probabilities. Also should document how ties (duplicate streamflow values are handled). By default, numpy.percentile does not handle tied values in any way (at least as far as I can tell). Whereas, the exceedance calculation implicitly is using a min_rank type calculation through np.sum(values_to_compare >= x) where duplicate x values receive the same exceedance probability.

Should look and double check how calculation is done in WaterWatch, NWIS Stats Service, or SWToolbox