Some of the SI.Posterior Samples are huge

[jstockwin]

See https://github.com/jstockwin/EpiEstimApp/tree/master/inst/app/datasets/SIPosteriorSamples and note how big the EcuadorRotavirus files for off1W and off1L are. Is this expected?

Tests are currently disabled for off1W, L, and off1L for this reason.

@robin-thompson Do you know if this is expected..?

[jstockwin]

Note this is probably the same as #66, but last time we were removing that dataset anyway so it wasn't an issue.

[jstockwin]

This happens when there is not much SI data, so it can choose ridiculous params for the distributions, meaning it has to go to very high values of time before the full distribution is small (i.e. max_time = inverse_CDF(0.999)).

This may cause problems, but there's not much we can do, short of lowering the 0.999 tolerance, which is here: https://github.com/annecori/EpiEstim/blob/hackout3/R/coarse2estim.R#L74

[jstockwin]

OR we can change coarse2estim so that instead of forcing every column to sum to >0.999, stop when the probability that the serial interval is more than t (over all columns) is less than 0.001.

@robin-thompson Is going to work out how to do this.