OCR tool suggestions

As mentioned in #336, i think the OCR tool could use some QOL improvments :

• the OCR default language is always english, which may give bad results if you're scanning pages written in other languages. Either detect the current language from the browser to apply automatically as the default one, or ask the user before the first page is processed. Alternatively, the OCR tool could automatically detect the scanned language (upstream improvement).
• offer pdf export format
• have the ability to scan several pages one after the other to export into a single zip file or pdf.
• improve efficiency of double page scans (upstream improvement), or warn that double page scans won't work well.

Thanks a lot for the work you've already done with Omnitools, it's really great !!