Empirical p-value calculation

[biodataganache]

I'd like to have a generalized framework for calculation of empirical p-values from all the enrichment methods. There are multiple ways to accomplish this - but I think a simple and universal method is to randomize gene labels within pathway designations (that is, keep the same number in each pathway, but randomly reassign genes to different pathways). I think this could be accomplished at the leapR.R .enrichment_wrapper function. Given an optional number of background randomizations the code could simply run the same enrichment multiple times with different randomized gene sets and then compare to the real enrichment results. Output could be a results table (same format as original) where the p-value is now empirically calculated.