feat(learning-harness): absence rows travel with their registration

Eight of the ten vendored access-control rows had no named handler at all,
so every one of them would have scored `missing_context` forever rather
than as a detector miss. After the re-harvest nine are named and the tenth
states its reason.

The handler-context mode now carries the class its framework registers,
multi-line decorators, and commented-out guards -- `#@token_required`
above a ThreatByte handler is the entire bug. Where the registration lives
in another document (connexion's `operationId`, an express router), a
recipe names `context = [...]`, the harvester writes that stanza per side,
and the pair materializes both files at their repository-relative paths so
the entry-point mapper sees the layout the project has. `harvest_recipe`
fetches every blob before the first write and the driver takes `--report`,
so a failed fetch leaves a reviewed excerpt exactly as it was.

Two defects the re-harvest exposed: `.mjs` was not a known suffix, so the
one pair vendored as an ES module had no file target at all; and
`_python_route_access` matched auth words inside decorator prose, reading
`@ns.doc(description='... without proper authorization checks')` as a
guard. Recorded in the hygiene artifact: nine of the ten duplicate groups
sit on both sides of the split, which the repository-level split cannot
see.

Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
Claude-Session: https://claude.ai/code/session_01T7XD7x5KonMdmGRkp63EQG

benchmarks/pairs/catalog_gen.py

@@ -37,6 +37,14 @@ def excerpt_rel(recipe: dict[str, object]) -> tuple[str, str]:
     return f"{name}/vuln{ext}", f"{name}/fixed{ext}"
 
 
+def context_rel(recipe: dict[str, object]) -> list[tuple[str, str, str]]:
+    """`(relpath, vuln document, fixed document)` for every document that registers the handler from outside its module."""
+    name = str(recipe["name"])
+    value = recipe.get("context") or ()
+    relpaths = (value,) if isinstance(value, str) else tuple(str(item) for item in value)
+    return [(relpath, f"{name}/context/vuln/{relpath}", f"{name}/context/fixed/{relpath}") for relpath in relpaths]
+
+
 def _q(value: object) -> str:
     return str(value).replace("\\", "\\\\").replace('"', '\\"')
 
@@ -85,6 +93,14 @@ def catalog_text(slice_name: str, recipes: list[dict[str, object]]) -> str:
             lines.append(f'reviewer = "{_q(reviewer)}"')
         if recipe.get("known_limit"):
             lines.append(f'known_limit = "{_q(recipe["known_limit"])}"')
+        for relpath, context_vuln, context_fixed in context_rel(recipe):
+            lines += [
+                "",
+                "[[pair.context]]",
+                f'relpath = "{_q(relpath)}"',
+                f'vuln = "{_q(context_vuln)}"',
+                f'fixed = "{_q(context_fixed)}"',
+            ]
         lines += [
             "",
             "[[pair.expected]]",
@@ -95,6 +111,8 @@ def catalog_text(slice_name: str, recipes: list[dict[str, object]]) -> str:
         if recipe.get("function"):
             lines.append(f'function = "{_q(recipe["function"])}"')
         lines.append(f'mechanism = "{_q(recipe.get("mechanism", "other"))}"')
+        if recipe.get("family"):
+            lines.append(f'family = "{_q(recipe["family"])}"')  # Req 1: the family routes the work, never a CWE lookup
         if recipe.get("rule_id"):
             lines.append(f'rule_id = "{_q(recipe["rule_id"])}"')
         if recipe.get("sink"):