[Feature]: x86 container upgrade

[daohu527]

Use Case / Motivation

Orin has been upgraded to Jetpack 6.2, but it's based on L4T.

Proposed Implementation

x86 Docker-based generation support

Platform:
 - Distribution: Ubuntu 22.04 Jammy Jellyfish
 - Release: 5.15.148-tegra
Libraries:
 - CUDA: 12.6.68
 - cuDNN: 9.3.0.75
 - TensorRT: 10.3.0.30
 - VPI: 3.2.4
 - Vulkan: 1.3.204
 - OpenCV: 4.8.0 - with CUDA: NO

Alternatives Considered

No response

[daohu527]

We have upgraded the CUDA and Tensor versions to support 40-series graphics cards. #151

Environment

This release supports both ARM (Jetson) and x86 GPU platforms.

Platform	CUDA	TensorRT	Base
ARM (Jetson)	12.6	10.3	JetPack 6.2.1
x86 GPU	12.8.1	10.9.0.34	nvidia/cuda:12.8.1-cudnn-devel-ubuntu22.04

JetPack 6.2.1 provides CUDA 12.6 and TensorRT 10.3 as part of the Jetson AI stack

Note

• TensorRT engines are not cross-compatible between ARM and x86.
• Export models as ONNX and build TensorRT engines on the target platform.