fix(langchain): sort similarity_search() results by distance + tighten test

- Sort vector_search results client-side by r.distance (ascending) to
  match the public contract promised in the docstring
- Replace isinstance(results, list) with results == [] assertion to
  explicitly enforce the empty-vector guard contract

Author: polaz

langchain-coordinode/langchain_coordinode/graph.py

@@ -219,11 +219,14 @@ def similarity_search(
         """
         if not query_vector:
             return []
-        results = self._client.vector_search(
-            label=label,
-            property=property,
-            vector=query_vector,
-            top_k=k,
+        results = sorted(
+            self._client.vector_search(
+                label=label,
+                property=property,
+                vector=query_vector,
+                top_k=k,
+            ),
+            key=lambda r: r.distance,
         )
         return [{"id": r.node.id, "node": r.node.properties, "distance": r.distance} for r in results]
 

tests/integration/adapters/test_langchain.py

@@ -168,7 +168,7 @@ def test_similarity_search_returns_results(graph, unique_tag):
 def test_similarity_search_empty_vector_returns_empty(graph):
     """similarity_search() with an empty vector list returns an empty list without error."""
     results = graph.similarity_search([], k=5)
-    assert isinstance(results, list)
+    assert results == []
 
 
 def test_schema_refreshes_after_add(graph, unique_tag):