csv_to_wfdb docstring has incorrect parameter descriptions #594

Found while reviewing #591, which fixes the summary, record_only, write_dir and return value descriptions. Some other parameter descriptions in the csv_to_wfdb docstring are wrong, mostly because they were copied from elsewhere:

• file_name is copied from rdrecord. It says the name is given "without any file extensions" and refers to pn_dir, which csv_to_wfdb doesn't have. It should be the path to the CSV file.
• fs says "if it is missing, a value of 250 is assumed", but it is a required argument.
• Several other descriptions (e.g. baseline, skew, byte_offset, checksum) are copied from the header file specification and describe header syntax rather than the Python arguments.
• header says the default signal names are [ch_1, ch_2, ...], but the code produces ch_0, ch_1, ... (

  wfdb-python/src/wfdb/io/convert/csv.py

  Line 323 in b8914e1

  sig_name = ["ch_" + str(i) for i in range(n_sig)]

  ).