np.unique() sort fix for segments

[glenn-jocher]

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

🌟 Summary

Enhancement to duplicate label removal logic in dataset verification.

📊 Key Changes

• Improved the method for identifying and removing duplicate labels in datasets.
• Maintained integrity of the corresponding segmentations when duplicates are removed.

🎯 Purpose & Impact

• 🛠️ Purpose: The new code is designed to more accurately identify and remove duplicate labels, which can cause issues during model training.
• 📈 Impact: This change helps ensure that data used for training machine learning models is clean and error-free, potentially improving model accuracy and reliability for users.
• 🧹 Users can expect a more streamlined data cleaning process with reduced chances of encountering problems due to duplicate annotations in their datasets.