------------------
- Make meggie consistent across induced response channel averages: baseline is corrected to before averaging over channels
- Pin matplotlib==3.7.3 as newer matplotlibs are not compatible with pinned mne==1.3.1
- Fix bug where saving TFR's resulted in csv file for each subject even though there should be only one file with all subjects' data in.