Suppress autogenerated module docstrings from rationale extraction (#882)

Alembic/Flask-Migrate revisions, Django migrations, and protobuf/OpenAPI
generated files produce hundreds of degree-1 rationale nodes labeled as
'possible documentation gaps'. Their module docstrings are revision
annotations or boilerplate, not architectural rationale.

- Add _is_autogenerated_python() in extract.py detecting Alembic, Django
  migrations, and generic DO-NOT-EDIT markers; skip module docstring only
- Function/class docstrings inside those files still extracted as normal
- report.py: exclude file_type=rationale nodes from isolated-node gaps
  section — rationale nodes are degree-1 by construction; flagging them
  as missing edges was always wrong
- 5 new tests covering Alembic, Django, protobuf, false-positive guard,
  and function-docstring passthrough

graphify/extract.py

@@ -1866,6 +1866,27 @@ def _extract_generic(path: Path, config: LanguageConfig) -> dict:
 _RATIONALE_PREFIXES = ("# NOTE:", "# IMPORTANT:", "# HACK:", "# WHY:", "# RATIONALE:", "# TODO:", "# FIXME:")
 
 
+def _is_autogenerated_python(source: bytes) -> bool:
+    """Return True if this Python file is auto-generated and its module docstring is noise.
+
+    Covers: Alembic/Flask-Migrate revisions, Django migrations, protobuf/gRPC/OpenAPI stubs.
+    Module docstrings in these files are change annotations or boilerplate, not rationale.
+    """
+    head = source[:2048].decode("utf-8", errors="replace")
+    # Generic generated-file markers (protobuf, gRPC, OpenAPI codegen, etc.)
+    if any(m in head for m in ("DO NOT EDIT", "@generated", "Generated by the protocol buffer")):
+        return True
+    # Alembic / Flask-Migrate revision files
+    if (re.search(r"^revision\s*[:=]", head, re.MULTILINE)
+            and "def upgrade(" in head
+            and "down_revision" in head):
+        return True
+    # Django migrations
+    if "class Migration(migrations.Migration)" in head and "operations" in head:
+        return True
+    return False
+
+
 def _extract_python_rationale(path: Path, result: dict) -> None:
     """Post-pass: extract docstrings and rationale comments from Python source.
     Mutates result in-place by appending to result['nodes'] and result['edges'].
@@ -1924,10 +1945,13 @@ def _extract_python_rationale(path: Path, result: dict) -> None:
             "weight": 1.0,
         })
 
-    # Module-level docstring
-    ds = _get_docstring(root)
-    if ds:
-        _add_rationale(ds[0], ds[1], file_nid)
+    # Module-level docstring — skip for auto-generated files (Alembic, Django
+    # migrations, protobuf stubs, etc.) whose module docstrings are revision
+    # annotations, not architectural rationale.
+    if not _is_autogenerated_python(source):
+        ds = _get_docstring(root)
+        if ds:
+            _add_rationale(ds[0], ds[1], file_nid)
 
     # Class and function docstrings
     def walk_docstrings(node, parent_nid: str) -> None:

graphify/report.py

@@ -164,7 +164,10 @@ def generate(
 
     isolated = [
         n for n in G.nodes()
-        if G.degree(n) <= 1 and not _is_file_node(G, n) and not _is_concept_node(G, n)
+        if G.degree(n) <= 1
+        and not _is_file_node(G, n)
+        and not _is_concept_node(G, n)
+        and G.nodes[n].get("file_type") != "rationale"
     ]
     thin_communities = {
         cid: nodes for cid, nodes in communities.items()