Adds various functionality for supporting other OpenAI compatible models: allows running localization and repair over multiple threads (which is very useful for slower AI services), resuming localization if it gets interrupted, reducing context length when it would overflow the context window, separate logging files for each instance to make log files more readable, ect.
Adds various functionality for supporting other OpenAI compatible models: allows running localization and repair over multiple threads (which is very useful for slower AI services), resuming localization if it gets interrupted, reducing context length when it would overflow the context window, separate logging files for each instance to make log files more readable, ect.