Issues with smoothing signal branch

[dillancellier]

• Fix it so that the filtering is done on the uncropped signal
• Also there is an error when the gaus param bounds don't match the filtered signal, use this code to recreate:

time_range = (0,2)
baseline = (-3,0)
ep_init = {'peak_width_limits': (0.1, 2), 'max_n_peaks': 3, 
 'min_peak_height': 0, 'peak_threshold': 2, 
 'peak_mode': 'gaussian', 'gauss_overlap_thresh': 0.75, 'amplitude_fraction': 0.5, 'filter_signal':True}

times = np.load(file='C:\\Users\\dillc\\OneDrive\\UCSD\\Voytek_Lab\\ERPparamPublication\\notebooks\\development\\toolbox\\error_times5.npy')
signal = np.load(file='C:\\Users\\dillc\\OneDrive\\UCSD\\Voytek_Lab\\ERPparamPublication\\notebooks\\development\\toolbox\\error_signal5.npy')
ep = ERPparam(**ep_init)
ep.fit(time=times, signal=signal, time_range=time_range, baseline=baseline)
ep.plot()