feat: tesseract implementation

[Claiyc]

PR description

Describe your changes in detail here
Once merged, this will add image-to-text translation using tesseract.js (not using regex recognition yet).
This will take capture areas into consideration (only extract text in capture areas)

As of now this will just print the recognized text into the run-view.

Definition Of Done (DoD)

This PR can be squashed / merged if

• a developer is assigned
• the PR is NOT estimated
• the PR is labeled
• the PR is assigned to the current sprint
• a meaningful title has been set according to https://www.conventionalcommits.org/
• the PR is described in detail
• the PR links to an issue
• the PR has been reviewed

Add additional conditions here if necessary for this PR
fixes #23

[Claiyc]

First part of implementation. Currently works as "call once, run once" -> should of course work as background-loop in the end. Mapping of capture areas to actual stream is not correct yet.
As of now it's quite the ugly chaining of async functions.. Will probably have to iterate on that.

[tonoizer]

Over all, the implementation looks quite good, but we will see how good this works together with the regex matcher.

[tonoizer]

Looks good to me.