AMD Quark for ONNX Model Support

The following models and techniques are demonstrated in the AMD Quark for ONNX examples:

Quantization Technique	Models
Accuracy improvement¹	densenet121, Llama2-7B, mobilenetv2_050, opt-125m, resnet50-v1-12, timm/resnet152
Auto-Search	Yolov3
Dynamic Quantization	Llama-2-7b, opt-125m
Hugging Face TIMM Quantization	mobilenetv2_100, various
Image Classification	Resnet50-v1-12
Language Model Quantization	opt-125m
Auto-Search for AMD Ryzen AI	Resnet50-v1-12, yolo_nas_s, yolov3
Post Training Quantization (PTQ) for AMD Ryzen AI	Resnet50-v1-12
Object Detection PTQ for AMD Ryzen AI	yolov8n-face

¹Accuracy improvement includes examples with: AdaQuant, AdaRound, Block Floating Point (BFP), Cross-Layer Equalization (CLE), GPTQ, Mixed Precision, Microscaling (MX) data types, QuaRot, and SmoothQuant.