Remove redundant metadata (#685)

The metadata field (list of triples) in the pipeline Metadata class
was redundant. Document metadata triples already flow directly from
librarian to triple-store via emit_document_provenance() - they don't
need to pass through the extraction pipeline.

Additionally, chunker and PDF decoder were overwriting metadata to []
anyway, so any metadata passed through the pipeline was being
discarded.

Changes:
- Remove metadata field from Metadata dataclass
  (schema/core/metadata.py)
- Update all Metadata instantiations to remove metadata=[]
  parameter
- Remove metadata handling from translators (document_loading,
  knowledge)
- Remove metadata consumption from extractors (ontology, agent)
- Update gateway serializers and import handlers
- Update all unit, integration, and contract tests

tests/unit/test_knowledge_graph/conftest.py

@@ -29,11 +29,10 @@ class Triple:
         self.o = o
 
 class Metadata:
-    def __init__(self, id, user, collection, metadata):
+    def __init__(self, id, user, collection):
         self.id = id
         self.user = user
         self.collection = collection
-        self.metadata = metadata
 
 class Triples:
     def __init__(self, metadata, triples):
@@ -110,7 +109,6 @@ def sample_triples(sample_triple):
         id="test-doc-123",
         user="test_user",
         collection="test_collection",
-        metadata=[]
     )
     
     return Triples(
@@ -126,7 +124,6 @@ def sample_chunk():
         id="test-chunk-456",
         user="test_user",
         collection="test_collection",
-        metadata=[]
     )
     
     return Chunk(

tests/unit/test_knowledge_graph/test_agent_extraction.py

@@ -51,13 +51,6 @@ class TestAgentKgExtractor:
         """Sample metadata for testing"""
         return Metadata(
             id="doc123",
-            metadata=[
-                Triple(
-                    s=Term(type=IRI, iri="doc123"),
-                    p=Term(type=IRI, iri="http://example.org/type"),
-                    o=Term(type=LITERAL, value="document")
-                )
-            ]
         )
 
     @pytest.fixture
@@ -274,7 +267,7 @@ This is not JSON at all
 
     def test_process_extraction_data_no_metadata_id(self, agent_extractor):
         """Test processing when metadata has no ID"""
-        metadata = Metadata(id=None, metadata=[])
+        metadata = Metadata(id=None)
         data = [
             {"type": "definition", "entity": "Test Entity", "definition": "Test definition"}
         ]
@@ -345,8 +338,6 @@ This is not JSON at all
         assert sent_triples.metadata.id == sample_metadata.id
         assert sent_triples.metadata.user == sample_metadata.user
         assert sent_triples.metadata.collection == sample_metadata.collection
-        # Note: metadata.metadata is now empty array in the new implementation
-        assert sent_triples.metadata.metadata == []
         assert len(sent_triples.triples) == 1
         assert sent_triples.triples[0].s.iri == "test:subject"
 
@@ -371,8 +362,6 @@ This is not JSON at all
         assert sent_contexts.metadata.id == sample_metadata.id
         assert sent_contexts.metadata.user == sample_metadata.user
         assert sent_contexts.metadata.collection == sample_metadata.collection
-        # Note: metadata.metadata is now empty array in the new implementation
-        assert sent_contexts.metadata.metadata == []
         assert len(sent_contexts.entities) == 1
         assert sent_contexts.entities[0].entity.iri == "test:entity"
 

tests/unit/test_knowledge_graph/test_agent_extraction_edge_cases.py

@@ -177,13 +177,13 @@ class TestAgentKgExtractionEdgeCases:
             pass
 
         # Test with metadata without ID
-        metadata = Metadata(id=None, metadata=[])
+        metadata = Metadata(id=None)
         triples, contexts = agent_extractor.process_extraction_data([], metadata)
         assert len(triples) == 0
         assert len(contexts) == 0
 
         # Test with metadata with empty string ID
-        metadata = Metadata(id="", metadata=[])
+        metadata = Metadata(id="")
         data = [{"type": "definition", "entity": "Test", "definition": "Test def"}]
         triples, contexts = agent_extractor.process_extraction_data(data, metadata)
 
@@ -193,7 +193,7 @@ class TestAgentKgExtractionEdgeCases:
 
     def test_process_extraction_data_special_entity_names(self, agent_extractor):
         """Test processing with special characters in entity names"""
-        metadata = Metadata(id="doc123", metadata=[])
+        metadata = Metadata(id="doc123")
 
         special_entities = [
             "Entity with spaces",
@@ -225,7 +225,7 @@ class TestAgentKgExtractionEdgeCases:
 
     def test_process_extraction_data_very_long_definitions(self, agent_extractor):
         """Test processing with very long entity definitions"""
-        metadata = Metadata(id="doc123", metadata=[])
+        metadata = Metadata(id="doc123")
 
         # Create very long definition
         long_definition = "This is a very long definition. " * 1000
@@ -247,7 +247,7 @@ class TestAgentKgExtractionEdgeCases:
 
     def test_process_extraction_data_duplicate_entities(self, agent_extractor):
         """Test processing with duplicate entity names"""
-        metadata = Metadata(id="doc123", metadata=[])
+        metadata = Metadata(id="doc123")
 
         data = [
             {"type": "definition", "entity": "Machine Learning", "definition": "First definition"},
@@ -269,7 +269,7 @@ class TestAgentKgExtractionEdgeCases:
 
     def test_process_extraction_data_empty_strings(self, agent_extractor):
         """Test processing with empty strings in data"""
-        metadata = Metadata(id="doc123", metadata=[])
+        metadata = Metadata(id="doc123")
 
         data = [
             {"type": "definition", "entity": "", "definition": "Definition for empty entity"},
@@ -291,7 +291,7 @@ class TestAgentKgExtractionEdgeCases:
 
     def test_process_extraction_data_nested_json_in_strings(self, agent_extractor):
         """Test processing when definitions contain JSON-like strings"""
-        metadata = Metadata(id="doc123", metadata=[])
+        metadata = Metadata(id="doc123")
 
         data = [
             {
@@ -315,7 +315,7 @@ class TestAgentKgExtractionEdgeCases:
 
     def test_process_extraction_data_boolean_object_entity_variations(self, agent_extractor):
         """Test processing with various boolean values for object-entity"""
-        metadata = Metadata(id="doc123", metadata=[])
+        metadata = Metadata(id="doc123")
 
         data = [
             # Explicit True
@@ -343,7 +343,7 @@ class TestAgentKgExtractionEdgeCases:
     @pytest.mark.asyncio
     async def test_emit_empty_collections(self, agent_extractor):
         """Test emitting empty triples and entity contexts"""
-        metadata = Metadata(id="test", metadata=[])
+        metadata = Metadata(id="test")
         
         # Test emitting empty triples
         mock_publisher = AsyncMock()
@@ -389,7 +389,7 @@ class TestAgentKgExtractionEdgeCases:
 
     def test_process_extraction_data_performance_large_dataset(self, agent_extractor):
         """Test performance with large extraction datasets"""
-        metadata = Metadata(id="large-doc", metadata=[])
+        metadata = Metadata(id="large-doc")
 
         # Create large dataset in JSONL format
         num_definitions = 1000

tests/unit/test_knowledge_graph/test_object_extraction_logic.py

@@ -314,7 +314,6 @@ class TestObjectExtractionBusinessLogic:
             id="test-extraction-001",
             user="test_user", 
             collection="test_collection",
-            metadata=[]
         )
         
         values = [{

tests/unit/test_knowledge_graph/test_triple_construction.py

@@ -373,7 +373,6 @@ class TestTripleConstructionLogic:
             id="test-doc-123",
             user="test_user",
             collection="test_collection",
-            metadata=[]
         )
 
         # Act