feat: Implement knowledge import feature for characters

- Added KnowledgeImport page for importing character knowledge from URLs.
- Integrated URL validation and error handling for unsupported websites.
- Created API endpoints for importing content from URLs and retrieving character knowledge.
- Enhanced VectorStoreService with logging and error handling for vector memory storage.
- Updated frontend to display knowledge sources and manage them effectively.
- Added support for fetching recent character knowledge as a fallback in similarity searches.
- Updated OpenAPI documentation to reflect new import functionality.

.env.example

@@ -20,6 +20,24 @@ EMBEDDING_MODEL=Xenova/all-MiniLM-L6-v2
 EMBEDDING_DIMENSION=384
 EMBEDDING_DEVICE=cpu
 
+# Use quantized model for lower memory usage (~4x smaller, slightly less accurate)
+# Set to 'true' for low-memory systems (512MB-1GB RAM)
+EMBEDDING_QUANTIZED=false
+
+# Node.js Memory Configuration (increase if embedding causes OOM)
+# For 512MB RAM VPS: NODE_OPTIONS=--max-old-space-size=384
+# For 1GB RAM VPS: NODE_OPTIONS=--max-old-space-size=768
+# For 2GB RAM VPS: NODE_OPTIONS=--max-old-space-size=1536
+# Default (no env var): Node uses ~4GB or system limit
+#NODE_OPTIONS=--max-old-space-size=768
+
+# Request Logging Configuration
+# Enable/disable request logging (default: true)
+#REQUEST_LOGGER=true
+
+# Log level: verbose (detailed), standard (default), minimal (status only)
+#REQUEST_LOGGER_LEVEL=standard
+
 # HuggingFace API (optional - if not using local embeddings)
 # HUGGINGFACE_API_KEY=hf_...