Issue: Some PDF samples are parsed as empty.

[ygw365]

2407.19456v1.pdf
Ominiparse is very powerful and can parse various types of files quite well. However, some PDF samples return empty results, as if the text inside them does not exist.

this is response

{"text":"","images":[],"metadata":{"languages":["en"],"filetype":"other"},"chunks":[]}