Fix pre_process method to convert NumPy array to PyTorch tensor

[Sachin-NK]

📝 Description

This pull request fixes the pre_process method in the TorchInferencer to correctly handle NumPy arrays before performing PyTorch operations. The main changes include:

• Adding a batch dimension to the input image if it has 3 dimensions.
• Converting the NumPy array to a PyTorch tensor.
• Moving the tensor to the specified device.

🛠️ Fixes #2452

✨ Changes

[ ] 🐞 Bug fix (non-breaking change which fixes an issue)

• Updated pre_process method to add a batch dimension using np.expand_dims if the input image has 3 dimensions.
• Converted the input image from a NumPy array to a PyTorch tensor using torch.from_numpy.
• Moved the tensor to the specified device using the to method.

✅ Checklist

• 📋 I have summarized my changes in the CHANGELOG and followed the guidelines for my type of change (skip for minor changes, documentation updates, and test enhancements).
• 📚 I have made the necessary updates to the documentation (if applicable).
• 🧪 I have written tests that support my changes and prove that my fix is effective or my feature works (if applicable).

[Sachin-NK]

Fixed the issue by converting image to a PyTorch tensor before calling unsqueeze(). When you have some time, please take a look and let me know if any further adjustments are needed.