Written Number Parsing Not as Expected

[darakelian]

Based on the description of the library and the various examples, I had assumed English spelling of words such as "twenty", "thirty", "forty", etc. would be properly parsed by this library. Upon investigating the code, it seems that for English, this library only parses the spelled out numbers 1-12: https://github.com/scrapinghub/dateparser/blob/master/dateparser/data/date_translation_data/en.py#L789 which honestly was a surprise. Is there any way to support parsing something like "in twenty minutes"? Would you guys be open to a PR adding this extra support?