feat(stage6): source independence & citation graph — detect syndication, shared origins, text similarity

- SourceIndependenceModel: per-source independence_score (0.0-1.0), syndication_group_id, primary_source_id, content_hash, shared_urls
- CitationGraphEdgeModel: directed edges (SYNDICATED, QUOTES, LINKS_TO, REPOST, SIMILAR_CONTENT) with confidence + evidence
- Content-Hash (SHA-256): instant syndication detection for identical content
- difflib Vorfilterung: >60% → LLM, >80% → high confidence, 100% → immediate syndication
- LLM-Pairwise-Analysis: two-text-comparison for suspicious pairs only (bounded concurrency)
- independence_score: 1.0 base, -0.4 for syndicated, -0.1 per high-similarity pair
- Pydantic schemas: SourceIndependenceScore, CitationGraphEdge, SyntacticSimilarityResult, LlmSyndicationAnalysis, SourceIndependenceAnalysisResult
- LLM response parser: handles JSON, markdown code blocks, partial/invalid JSON
- 43 tests: content hash, similarity thresholds, LLM parsing, analyzer integration, edge cases, prompt templates

src/nsct/storage/models.py

@@ -16,6 +16,7 @@ from sqlalchemy import (
     ForeignKey,
     Index,
     Integer,
+    JSON,
     String,
     Text,
 )
@@ -58,6 +59,16 @@ class EdgeRelation(str, enum.Enum):
     SUPPLEMENTS = "supplements"
 
 
+class CitationEdgeType(str, enum.Enum):
+    """Kanten-Typen im Source-Citation-Graph (Stage 6: Source Independence)."""
+
+    SYNDICATED = "syndicated"
+    QUOTES = "quotes"
+    LINKS_TO = "links_to"
+    REPOST = "repost"
+    SIMILAR_CONTENT = "similar_content"
+
+
 # ---------------------------------------------------------------------------
 # Base
 # ---------------------------------------------------------------------------
@@ -112,6 +123,12 @@ class SourceModel(Base):
 
     # Relationships
     claims = relationship("ClaimModel", back_populates="source", cascade="all, delete-orphan")
+    independence = relationship(
+        "SourceIndependenceModel",
+        back_populates="source",
+        uselist=False,
+        cascade="all, delete-orphan",
+    )
 
     __table_args__ = (
         Index("ix_sources_domain", "domain"),
@@ -167,7 +184,7 @@ class EvidenceRelationModel(Base):
 
 
 # ---------------------------------------------------------------------------
-# CitationEdge
+# CitationEdge (Stage 0 — legacy claim-level edges)
 # ---------------------------------------------------------------------------
 
 
@@ -186,6 +203,55 @@ class CitationEdgeModel(Base):
     )
 
 
+# ---------------------------------------------------------------------------
+# SourceIndependence (Stage 6 — per-source independence metadata)
+# ---------------------------------------------------------------------------
+
+
+class SourceIndependenceModel(Base):
+    """Speichert den Independence-Score und Syndication-Informationen pro Quelle (Stage 6)."""
+
+    __tablename__ = "source_independence"
+
+    id = Column(String(36), primary_key=True, default=lambda: str(uuid4()))
+    source_id = Column(String(36), ForeignKey("sources.id"), nullable=False, unique=True)
+    independence_score = Column(Float, nullable=False, default=1.0)
+    syndication_group_id = Column(String(36), nullable=True)
+    primary_source_id = Column(String(36), ForeignKey("sources.id"), nullable=True)
+    content_hash = Column(String(64), nullable=True)
+    text_similarity_high = Column(Integer, nullable=False, default=0)
+    shared_urls = Column(JSON, nullable=False, default=dict)
+    created_at = Column(DateTime, nullable=False, default=datetime.utcnow)
+    updated_at = Column(DateTime, nullable=False, default=datetime.utcnow)
+
+    # Relationships
+    source = relationship("SourceModel", back_populates="independence")
+
+
+# ---------------------------------------------------------------------------
+# CitationGraphEdge (Stage 6 — source-to-source edges)
+# ---------------------------------------------------------------------------
+
+
+class CitationGraphEdgeModel(Base):
+    """Directed edge zwischen zwei Quellen im Source-Citation-Graph (Stage 6)."""
+
+    __tablename__ = "citation_graph_edges"
+
+    id = Column(String(36), primary_key=True, default=lambda: str(uuid4()))
+    source_id = Column(String(36), ForeignKey("sources.id"), nullable=False)
+    target_source_id = Column(String(36), ForeignKey("sources.id"), nullable=False)
+    edge_type = Column(Enum(CitationEdgeType), nullable=False)
+    confidence = Column(Float, nullable=False, default=1.0)
+    evidence = Column(JSON, nullable=False, default=dict)
+    created_at = Column(DateTime, nullable=False, default=datetime.utcnow)
+
+    __table_args__ = (
+        Index("ix_citation_graph_edge_source_id", "source_id"),
+        Index("ix_citation_graph_edge_target_id", "target_source_id"),
+    )
+
+
 # ---------------------------------------------------------------------------
 # ResearchReport
 # ---------------------------------------------------------------------------