[BUG] processEMG.m Inappropriate EMG Normalization

[bfregly]

Describe the bug
The function processEMG.m which is used in preprocessing normalizes EMG data to maximum value over the current motion cycle. This normalization means that the same EMG channel will be normalized by different values for different motion trials. We need to remove the EMG normalization line

% Normalize by maximum value for each channel
emgData = emgData ./ max(emgData, [], 2);

and allow the user to normalize EMG data to maximum value over MULTIPLE trials using any desired combination of processed EMG trials.

To Reproduce
Nothing to reproduce - just need to remove one line of code.

Expected behavior
Output EMG data will not be normalized

Screenshots
If applicable, add screenshots to help explain your problem.

Desktop (please complete the following information):

• OS: All
• OpenSim Version: Any
• MATLAB Version: Any
• NMSM Pipeline Version: 1.0beta

Additional context
Add any other context about the problem here.

[cvhammond]

processEMG.m should be run before splitIntoTrials.m so it normalizes over the entire motion not over the individual motion cycle.

[SpencerTWilliams]

This came up again recently. The solution will be to normalize EMG over all trials after importing data into MTP. MTP activations will then be the expected magnitudes for NCP.