Update dataloaders.py

[glenn-jocher]

General cleanup and refactoring

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

🌟 Summary

Improved code simplicity and robustness in file management and image processing functions of the Ultralytics YOLOv5 repo.

📊 Key Changes

• Utilized contextlib.suppress to streamline handling of exceptions in exif_size, removing verbose try-except blocks.
• Simplified code in the flatten_recursive function by eliminating the redundant create_folder method.
• Updated string paths to use formatted strings (f-strings) for better readability and efficiency.
• Streamlined the _unzip function by reordering checks and assertions for clarity.

🎯 Purpose & Impact

• Enhances code readability and maintainability, making it easier for developers to navigate and contribute.
• Reduces the risk of errors by having less complex exception handling and more straightforward file operations.
• The use of modern Python features (context managers and formatted strings) aligns the code with contemporary best practices.
• Users can expect smoother operation due to the refined error handling and more robust file directory manipulations.