extcite v0.4.0

• bump serrano version (#55)
• bump bibtex-ruby version (#58)
• bump thor version (#59)
• bump faraday_middleware version (#60)
• updated extract_text_one method to use Timeout.timeout while looping through pages of a PDF to extract text - sometimes pdf-reader hangs, so only allow 1 second max before skipping to the next page
• add error catching to extract_from_metadata method for xml parsing of PDF metadata; sometimes PDF metadata is full of non-parseable XML