Poll for users who work in languages other than English

OCRmyPDF currently assumes all documents are English unless told otherwise with the -l argument.

For users who work in another language, would you generally prefer OCRmyPDF to assume that the locale set by the operating system is the one you usually work in? Is this how you assumed it would work?

[willaaam]

I always give the language anyway, I would find it kind of annoying for it to default to the locale, now behavior is more consistent should someone be running this in containers or something and forgetting to set the language param explicitly.

[stweil]

Ideally all compiled default values could be overridden by optional system and user default values – not only the default language. That’s a feature which is still missing in Tesseract.