Add TensorRT dependencies

[Zengyf-CVer]

When executing the command python export.py --weights yolov5s.pt --include engine --device 0, you need to install nvidia-pyindex and nvidia-tensorrt to export yolov5s.engine .

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

🌟 Summary

Enhanced support for NVIDIA TensorRT in YOLOv5 by updating dependencies.

📊 Key Changes

• Added nvidia-pyindex as a new requirement.
• Added nvidia-tensorrt as a new requirement.

🎯 Purpose & Impact

• Purpose: These changes aim to facilitate the export of YOLOv5 models to TensorRT format, which is designed for efficient inference on NVIDIA GPUs.
• Impact: Users looking to deploy YOLOv5 on NVIDIA hardware can now do so more seamlessly, potentially improving the performance and speed of their models in production. 🚀

[glenn-jocher]

@Zengyf-CVer PR is merged. Thank you for your contributions to YOLOv5 🚀 and Vision AI ⭐