Update to use correct HF dataset: modelbiome/ai_ecosystem

- Changed from ai_ecosystem_withmodelcards to ai_ecosystem
- This enables access to full 1.86M+ model dataset

backend.log

@@ -0,0 +1,6 @@
+INFO:     Will watch for changes in these directories: ['/Users/hamidaho/hf_viz/backend']
+INFO:     Uvicorn running on http://0.0.0.0:8000 (Press CTRL+C to quit)
+INFO:     Started reloader process [24332] using WatchFiles
+INFO:     Started server process [24386]
+INFO:     Waiting for application startup.
+Repo card metadata block was not found. Setting CardData to empty.

backend/services/model_tracker_improved.py

@@ -259,7 +259,7 @@ class ImprovedModelCountTracker:
             logger.error(f"Error extracting count from HF models page: {e}", exc_info=True)
             return None
     
-    def get_count_from_dataset_snapshot(self, dataset_name: str = "modelbiome/ai_ecosystem_withmodelcards") -> Optional[Dict]:
+    def get_count_from_dataset_snapshot(self, dataset_name: str = "modelbiome/ai_ecosystem") -> Optional[Dict]:
         """
         Alternative method: Get count from dataset snapshot (like ai-ecosystem repo does).
         This is faster but may be slightly outdated.

backend/utils/data_loader.py

@@ -10,7 +10,7 @@ import numpy as np
 class ModelDataLoader:
     """Load and preprocess model data from Hugging Face dataset."""
     
-    def __init__(self, dataset_name: str = "modelbiome/ai_ecosystem_withmodelcards"):
+    def __init__(self, dataset_name: str = "modelbiome/ai_ecosystem"):
         self.dataset_name = dataset_name
         self.df: Optional[pd.DataFrame] = None