Update framework to ONNX Runtime (FP32/FP16), remove Apple dependencies, add validation script for ONNX conversion with FP32-preserving ops, fix FP16 precision issues, update inference CLI with depth exaggeration, rename docs, and enable LFS support.

Refactor FP16 quantization using ONNX-native methods, update tolerance configs for depth/quaternions/colors, add FP32-preserving op block list, fix calibration workflow, enhance validation with FP16/FP32 distinction, optimize inference with external data support.

Fix precision tolerances, remove legacy FP16 logic, update data handling, standardize execution provider

Add precision constraints for ops like Resize/Gather, optimize FP16 casting with onnxoptimizer fallback, implement dynamic cast helpers, update CoreML provider priority, improve mixed-float error handling

inference script