- Added note in README about handling multiline text for language detection
- Updated example outputs in README for clarity
- Added docstrings to infer.py functions for better documentation
- Added test for exception handling in detect function in test_detect.py

- Refactor `get_model_loaded` to `load_model` with enhanced error handling
- Rename constants for better clarity and consistency
- Simplify language detection functions and improve docstrings
- Use `Optional` typing and rename parameters for clarity

Refactor model loading to use an enum and cache class.
Improve error handling and logging.
Add functionality to fallback to a local small model if download fails.

add `USE_STRICT_MODE` setting to enable/disable strict mode in detection
updated README.md to reflect the new setting
removed redundant `USE_STRICT_MODE` declaration in `infer.py`

- Updated README with new `use_strict_mode` parameter examples
- Removed unused imports and settings in `infer.py`
- Implemented `use_strict_mode` in model loading functions
- Added `NOTICE.MD` with license information

- Adjusted parameter alignment in `load_model` definition
- Refactored `load_large_model` logic for clarity
- Enhanced fallback logic for strict and non-strict modes

- Adjusted parameter alignment in `load_model` definition
- Refactored `load_large_model` logic for clarity
- Enhanced fallback logic for strict and non-strict modes

(feat): Add small model `lid.176.ftz` to library resources, for offline use