feat(compressors): add OWLv2 bbox crop to HashPipeline and refactor image utilities

- Add Owlv2ForObjectDetection and Owlv2Processor imports to model_loader
- Refactor load_dino_model to return tuple of processor and model
- Rewrite generate_proposals_batch to group images by bbox count for efficient batching
- Add _normalize_single_bbox_list helper for bbox normalization
- Update verification.py to use new pipeline architecture with detect/segment/filter/crop steps

mini-nav/configs/config.yaml

@@ -1,35 +1,35 @@
 model:
-  dino_model: "facebook/dinov2-large"
-  compression_dim: 512
-  device: "cuda:3" # auto-detect GPU
-  sam_model: "facebook/sam2.1-hiera-large" # SAM model name
-  sam_min_mask_area: 100 # Minimum mask area threshold
-  sam_max_masks: 10 # Maximum number of masks to keep
-  sam_points_per_batch: 64
-  compressor_path: null # Path to trained HashCompressor weights (optional)
+    dino_model: "facebook/dinov2-large"
+    compression_dim: 512
+    device: "cuda:2" # auto-detect GPU
+    sam_model: "facebook/sam2.1-hiera-large" # SAM model name
+    sam_min_mask_area: 100 # Minimum mask area threshold
+    sam_max_masks: 10 # Maximum number of masks to keep
+    sam_points_per_batch: 64
+    compressor_path: null # Path to trained HashCompressor weights (optional)
 
 output:
-  directory: "./outputs"
+    directory: "./outputs"
 
 dataset:
-  dataset_root: "datasets/InsDet-FULL"
-  output_dir: "datasets/InsDet-FULL/Synthesized"
-  num_objects_range: [3, 8]
-  num_scenes: 1000
-  object_scale_range: [0.1, 0.4]
-  rotation_range: [-30, 30]
-  overlap_threshold: 0.3
-  seed: 42
+    dataset_root: "datasets/InsDet-FULL"
+    output_dir: "datasets/InsDet-FULL/Synthesized"
+    num_objects_range: [3, 8]
+    num_scenes: 1000
+    object_scale_range: [0.1, 0.4]
+    rotation_range: [-30, 30]
+    overlap_threshold: 0.3
+    seed: 42
 
 benchmark:
-  dataset:
-    source_type: "huggingface"
-    path: "uoft-cs/cifar10"
-    img_column: "img"
-    label_column: "label"
-  task:
-    name: "recall_at_k"
-    type: "retrieval"
-    top_k: 1
-  batch_size: 64
-  model_table_prefix: "benchmark"
+    dataset:
+        source_type: "huggingface"
+        path: "uoft-cs/cifar10"
+        img_column: "img"
+        label_column: "label"
+    task:
+        name: "recall_at_k"
+        type: "retrieval"
+        top_k: 1
+    batch_size: 64
+    model_table_prefix: "benchmark"