Refactor `_readfile` to have a seperate `read_td_file` method

The lines https://gitlab.com/octopus-code/postopus/-/blob/main/src/postopus/files/pandas_text.py#L118 in _readfile method could be refactored to moves the code for extracting metadata from the file into a separate method called read_td_file. The method takes no parameters and returns a tuple containing the metadata dictionary, the column headers list, and the units list.

def read_td_file(self):
    with open(self.filepath, "r") as file:
        # skip "HEADER" line
        for i in range(2):
            _ = file.readline()

        # After HEADER and until # Iter, everything is metadata
        metadata_lines = 0
        metadata_dict = {}
        for line in file:
            if line.startswith("# Iter"):
                break
            metadata_lines += 1
            # split key and values if two spaces or more, ignore breaklines
            # and eq. signs
            split_line = re.split(
                r"\s{2,}", line[2:].replace("\n", "").replace("=", "")
            )
            metadata_dict[split_line[0]] = split_line[1:]

        # last two lines are column headers and units
        col_line = line
        unit_line = file.readline()

    # complicated splitting needed because some column titles have a
    # space inside
    column_headers = [
        s.strip() for s in col_line[1:-1].split("  ") if len(s) > 1
    ]

    # unit dict
    units = [s.strip() for s in unit_line[1:-1].split("  ") if len(s) > 1]

    return metadata_dict, column_headers, units