Add image-level split support for CLIP fine-tuning

Image-level splits allow the model to see some images from each logo
brand during training, unlike logo-level splits where test brands are
completely unseen. This is less rigorous but more representative of
real-world use.

Changes:
- Add configs/image_level_splits.yaml with gentler training settings:
  - split_level: "image" for image-level splits
  - temperature: 0.15 (softer contrastive learning)
  - learning_rate: 5e-6 (slower learning)
  - max_epochs: 30 (more epochs)

- Update training/dataset.py:
  - Add split_level parameter to LogoDataset
  - Implement _split_images() for image-level splitting
  - Update LogoContrastiveDataset to use split-specific image mappings

- Update training/config.py:
  - Add split_level field to TrainingConfig

- Update train_clip_logo.py:
  - Pass split_level to create_dataloaders

Usage:
  uv run python train_clip_logo.py --config configs/image_level_splits.yaml

training/config.py

@@ -20,7 +20,8 @@ class TrainingConfig:
     reference_dir: str = "reference_logos"
     db_path: str = "test_data_mapping.db"
 
-    # Data split ratios
+    # Data split configuration
+    split_level: str = "logo"  # "logo" for brand-level, "image" for image-level
     train_split: float = 0.7
     val_split: float = 0.15
     test_split: float = 0.15