Support for non-Latin characters

[robertknight]

ocrs should support models that can recognize non-Latin text.

Essential sub-tasks:

• Eliminate the hard-coded alphabet from the recognition process. IIRC this was from an EasyOCR model that I used at one point. (feat: customizable alphabet using OcrEngineParams #100)

Additional sub-tasks:

• Create a proof-of-concept recognition model that recognizes a non-Latin alphabet
• Create a standard file format for model metadata (including the alphabet) and support loading it. For example a config.json file can be shipped alongside the model and contains the alphabet.

[DehaiWang]

will support Chinese character？

[robertknight]

The goal is to make this possible. There are a lot of details still to be figured out.

[robertknight]

Eliminating the hard-coded alphabet from the recognition process. IIRC this was from an EasyOCR model that I used at one point.

This was completed in #100.