[Bug]: Inconsistent language order in tesseract calls

[abwiersma]

What were you trying to do?

I was trying to get consistent hocr results from a list of language models, but was finding that even though the list of languages supplied to ocrmypdf was consistent, the list of languages passed on to tesseract was randomly sorted.

So for example:
ocrmypdf -l lang1+lang2+lang3
would result in a random permutation of the -l parameter being passed on to tesseract, something like:
'tesseract', '-l', 'lang2+lang1+lang3

This breaks consistent language parsing as Tesseract has a sense of the primary language being given preference over the secondary languages.

Where are you installing from?

PyPI (pip, poetry, pipx, etc.)

What operating system are you working on?

Linux

Relevant log output

No response

[abwiersma]

Am writing a PR to fix this issue

[abwiersma]

Fixed with PR