feat: Add Knowledge Graph with entity extraction and graph traversal

[Nsuccess]

Closes #248

What this PR does

Adds a Knowledge Graph layer to MemU, enabling relationship-aware memory retrieval.

Features Added:

• KnowledgeGraph class using NetworkX for entity-relationship storage
• Entity extraction LLM prompt for extracting entities and relationships from text
• Graph traversal with get_subgraph_context() for context-aware retrieval
• Entity deduplication with fuzzy matching
• KnowledgeGraphMixin for easy service integration
• 26 unit tests covering all functionality

Files Added:

• src/memu/database/knowledge_graph.py - Core KnowledgeGraph class
• src/memu/prompts/entity_extraction.py - LLM prompt for entity extraction
• src/memu/app/knowledge_graph_mixin.py - Service mixin
• tests/test_knowledge_graph.py - 26 unit tests

Dependencies:

• Added networkx>=3.0 to pyproject.toml

[sairin1202]

Hi Nsuccess, thank you for the suggestion. At the moment, the knowledge graph approach conflicts with our current system architecture, so we’re unable to accept it. We appreciate your understanding.