Fix code formatting and test security setup

- Fix black formatting issues in Python files
- Add security test changes to README
- Test branch protection rules with test branch
- Verify security tools (safety, pip-audit) are working
- Found 3 vulnerabilities in dependencies (pip, starlette)
- Ready for PR creation to test full workflow

Author: AlmostBald-TRADING

src/contextforge_memory/__init__.py

@@ -1,4 +1,2 @@
 __all__ = ["__version__"]
 __version__ = "0.0.1"
-
-

src/contextforge_memory/main.py

@@ -94,14 +94,21 @@ def score(a: List[float], b: List[float]) -> float:
     with bucket.open("r", encoding="utf-8") as f:
         for line in f:
             data = orjson.loads(line)
-            if data.get("namespace") != req.namespace or data.get("project_id") != req.project_id:
+            if (
+                data.get("namespace") != req.namespace
+                or data.get("project_id") != req.project_id
+            ):
                 continue
             item_vec = _embed_text(data.get("text", ""))
             data["_score"] = score(query_vec, item_vec)
-            results.append(MemoryItem(**{k: v for k, v in data.items() if k != "_score"}))
+            results.append(
+                MemoryItem(**{k: v for k, v in data.items() if k != "_score"})
+            )
 
     # Sort by score descending based on cached scores; recompute on the fly for simplicity
-    results.sort(key=lambda it: -sum((x - y) ** 2 for x, y in zip(_embed_text(req.query), _embed_text(it.text))))
+    results.sort(
+        key=lambda it: -sum(
+            (x - y) ** 2 for x, y in zip(_embed_text(req.query), _embed_text(it.text))
+        )
+    )
     return SearchResponse(results=results[: max(1, req.top_k)])
-
-