This function reads downsample data values from the .rsk file and creates the
downsample field. The downsample data of a .rsk file is a summary of the dataset; its purpose is to give an overall impression of the data without needing to read in the full length, as this may be time and memory consuming. It is a downsample of the original dataset based on "Largest Triangle Three Buckets Algorithm" and contains up to 40960 samples per channel. For details of the algorithm, please refer to Downsampling time series for visual representation, Sveinn Steinarsson, 2013.
downsample will not exist when number of data samples per channel is smaller than 40960, in such case, we recommend users to view the data field directly.