fix: exclude correlation fields from the flex-auth request digest

Verified the digest join against flex-auth's T03 replay fixtures and found
request_digest was hashing fields docs/canonical-request-digest.md excludes.
The material is tenant, subject, action, resource, context only: id is
correlation, policy_version lives in provenance, caring_context is hashed
separately. This engine included all three when present.

Because the join adopts the served request id, every real production request
would have carried one, so the computed digest would have matched no issued
decision and failed closed against every correct allow. Same unsatisfiable
shape as the removed AUTHORITY constant.

The old pinned constant was computed with the id inside the material, so it
was wrong and its passing proved nothing. Replaced with fixture-driven tests
over two real envelopes (vendored with provenance) plus a structural test
that correlation fields do not move the digest. Both fixtures are needed:
input_claim_digests.context appears only with a non-empty context.

Also stops computing the native claim digest. The claim's binding.action and
binding.target speak approval-engine's vocabulary while ours speaks the
catalog's, and no mapping is published; flex-auth makes no cross-check and
states the correspondence is ours via pdp_digest. A claim recording no
pdp_digest now fails closed naming the missing mapping rather than comparing
two different languages. That mapping is a prerequisite for destroy.

274 tests pass. Production still fails closed.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01M65ovP3eiiPHubibvWs9mD

Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 393550@bnt-lap001
Assistant-Session: 4bb359f9-1f12-4410-9e76-079cf23c82e4

tests/test_consume_binding_join.py

@@ -77,7 +77,14 @@ def _served(entry=None, action="deactivate", fields=("api_token",), **over):
         "state": "valid",
         "valid_now": True,
         "consumed": False,
-        "binding": {**binding, "digest": claim_binding_digest(**binding)},
+        "binding": {
+            **binding,
+            "digest": claim_binding_digest(**binding),
+            # The issuer recorded the PDP digest at issue time. That is the only
+            # comparison usable today: the native digest speaks
+            # approval-engine's vocabulary and no mapping to ours is published.
+            "pdp_digest": request_digest(request),
+        },
         "freshness": {
             "observed_at": now.strftime("%Y-%m-%dT%H:%M:%SZ"),
             "ttl_seconds": 30,
@@ -182,3 +189,25 @@ def test_unreachable_approval_engine_fails_closed(tmp_path):
 def test_superseded_claim_fails_closed(tmp_path):
     with pytest.raises(DecisionError):
         _resolve(_Cfg(_token(tmp_path)), _entry(), _served(valid_now=False, reason_code="superseded"))
+
+
+def test_claim_without_pdp_digest_fails_closed_naming_the_missing_mapping(tmp_path):
+    """No published vocabulary mapping means the claim cannot be tied to this action.
+
+    approval-engine's binding.action/target use their vocabulary
+    ("secrets.kv.destroy", {"id":..., "stage":...}); ours uses the catalog's.
+    flex-auth makes no cross-check and states the correspondence is ours via
+    pdp_digest. Without one there is nothing sound to compare, so this must
+    refuse rather than fall back to comparing two different languages.
+    """
+    claim = _served()
+    claim["binding"].pop("pdp_digest")
+    with pytest.raises(DecisionError, match="no published mapping"):
+        _resolve(_Cfg(_token(tmp_path)), _entry(), claim)
+
+
+def test_wrong_pdp_digest_fails_closed(tmp_path):
+    claim = _served()
+    claim["binding"]["pdp_digest"] = "sha256:" + "c" * 64
+    with pytest.raises(DecisionError, match="pdp digest does not match"):
+        _resolve(_Cfg(_token(tmp_path)), _entry(), claim)