This expands the file types tested with file_search to include Word
documents (.docx), Markdown (.md), text (.txt), PDF (.pdf), and
PowerPoint (.pptx) files.
Python's mimetypes library doesn't actually recognize markdown docs as
text, so we have to handle that case specifically instead of relying
on mimetypes to get it right.
Signed-off-by: Ben Browning <bbrownin@redhat.com>
