Check ppf computation for hypergeometric and betabinomial

[aloctavodia]

See #407 and #411. For these two distributions, but not others, we have to add 1, so it seems we are missing something. Maybe we are making a mistake for the cdf computation. Although the test for the cdf pass.