Align power calculation in reports with ramulator method

[zduey]

Post-hoc reports use the full session of data to normalize the features, while ramulator uses the first four lists and then updates the mean/std by feature with each new list. This can lead to discrepancies in the classifier outputs calculated post hoc and those calculated during the session in real time. We have previously accepted this difference and assumed that it won't make a big difference, but it would certainly be better to be using the same procedure. This will also provide an additional check on ramulator