Draft: Add OCR feature to spectacle
Allows extracting the text from a newly created screenshot
Current issues:
- Language for extraction is hardcoded to english. We should detect the language available and allow the user to select which one they want to use.
- This feature is not optional (not sure this is really a problem)