Phi-3 Mii-4K-Istruct ONNX model for oxrutime-web

This is the same models as the official phi3 ox model with a few chages to make it work for oxrutime-web:

1. the model is fp16 with it4 block quatizatio for weights
2. the 'logits' output is fp32
3. the model uses MHA istead of GQA
4. ox ad exteral data file eed to stay below 2GB to be cacheable i chromium

Phi-3 Mini-4K-Instruct ONNX model for onnxruntime-web This is the same models as the official phi3 o