Compute Hartigan's dip test statistic for unimodality / multimodality and provide a test with simulation based p-values, where the original public code has been corrected.
