removed downsampling for high-pass filter in both scripts, thus scale 1 entropy is assessed for each frequency, with exclusive variations in the spectral content of the signal

added check for nans in input data, which can happen if coming from timelock with var trl lengths), which is converted to raw by ft_checkdata

added check for nans in input data, which can happen if coming from timelock with var trl lengths), which is converted to raw by ft_checkdata

% 181019 JQK | line 131, 249, 259, 261 287 commented
%            | line 163 verbosity changed
%            | save r parameter
%            | add option for HPF & BPF
%            | also use coarsegrainmethod for scale 1 (e.g., HPF)
% 180226 JQK | 364 ff. change encoding from sc to s to allow for nonlinear scale encoding
% 190130 JQK | filter entire time series first, then temporally segment
%            | do not require data.trialinfo, get trialsize from .trial
% 190405 JQK | integrated changes from previous scripts: adpative padding
%               length; no LPF at scale 1; changed indenting; removed
%               outdated code; corrected BP setting: at scale 1 only use HP
%               vs. LP