Qwen2-VL FP8/INT8 Quantization

[MrD005]

Implement FP8/INT8 quantization support for Qwen2-VL in TensorRT, optimizing LLM inference performance with reduced precision. Ensure compatibility, accuracy, and benchmarks for deployment scenarios.

[sunnyqgg]

Hi @MrD005 , for the FP8/INT8 LLM part, please refer to examples/qwen