In the event that we do not specify any filepaths to output this method's results, we should return a tuple containing both the edges significant based on the permutation test and the network itself.
In addition or alternatively, the CoNetwork class's to_dataframe method should have a parameter that allows for only the edges considered significant after the permutation test (each edge object would then have a significant boolean parameter) to be included in the dataframe. This would remove the need to pass in a whitelist parameter every time.