As evidenced by #116 , the "--language" flag isn't well known.

We should document it's usage so that people using Docsplit with foreign character sets do not have their document's UTF characters replaced by '?' by the TextCleaner.