feat(1e): HC3 additive generic + op/assertion split (orchestrator owns the op)

- orchestrator: per mutating tier, run optional pre-op seed hook (ops.py pre_<op>) → perform the op
  ONCE (harness-owned) → run generic assertion (unless opted out) AND overlay assertion, both against
  the shared post-op deployment. Op results passed op→assertion via run-scoped CCCI_OP_STATE_FILE.
- opt-out: CCCI_SKIP_GENERIC / CCCI_SKIP_GENERIC_<OP> / recipe_meta.SKIP_GENERIC (declarative).
- generic.py: split do_* into op primitives (perform_upgrade/backup/restore) + assertions
  (assert_upgraded/backup_artifact/restore_healthy) reading op_state(); deployed_identity now returns
  {version,image,chaos} (chaos label ready for HC1).
- generic test_<op>.py + all 6 recipe overlays migrated to assertion-only; pre-op seeding moved to
  per-recipe ops.py (pre_upgrade/pre_backup/pre_restore). install overlays unchanged (no op).
- deploy-count stays 1 (op primitives never call deploy_app). lint PASS; 8 unit tests PASS on cc-ci.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

tests/custom-html/test_backup.py

@@ -1,34 +1,21 @@
-"""custom-html — BACKUP overlay (Phase 1d, DG4): seed a known state, back it up (assert artifact),
-then mutate so the RESTORE overlay (test_restore.py) can prove the backed-up state returns. Runs on
-the shared deployment; the marker it leaves ("mutated") persists for the restore tier.
+"""custom-html — BACKUP overlay (Phase 1e HC3): assertion-only + additive.
 
-Reads the marker via `exec_in_app` (the file in the volume), NOT http: backup/restore preserve the
-VOLUME, and reading it directly is immune to the serving/container-routing race right after
-backup-bot-two cycles the app container (HTTP briefly served empty). Serving is proven separately by
-the install/upgrade tiers' assert_serving."""
+The orchestrator ran `ops.pre_backup` (seeded "original" into the served volume), then performed the
+backup ONCE. The generic backup tier already asserted a snapshot artifact was produced; this overlay
+ADDS the recipe-specific check: the seeded "original" state is intact in the volume post-backup
+(pre-mutation). The backup→restore divergence happens in `ops.pre_restore`. Reads via exec_in_app
+(volume-direct), immune to the post-backup serving race after backup-bot-two cycles the container."""
 
 import os
 import sys
 
 sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "..", "runner"))
-from harness import generic, lifecycle  # noqa: E402
+from harness import lifecycle  # noqa: E402
 
 MARKER_PATH = "/usr/share/nginx/html/ci-marker.txt"
 
 
-def _marker(domain: str) -> str:
-    return lifecycle.exec_in_app(domain, ["cat", MARKER_PATH]).strip()
-
-
-def test_backup_captures_state(live_app, meta):
-    domain = live_app
-    # 1) establish a known original state, then back it up (reuse the generic op: backup + assert a
-    #    snapshot artifact was produced)
-    lifecycle.exec_in_app(domain, ["sh", "-c", f"echo original > {MARKER_PATH}"])
-    assert _marker(domain) == "original"
-    snap = generic.do_backup(domain)
-    assert snap, "backup produced no snapshot artifact"
-
-    # 2) mutate state so a successful restore is observable (diverge from the backup)
-    lifecycle.exec_in_app(domain, ["sh", "-c", f"echo mutated > {MARKER_PATH}"])
-    assert _marker(domain) == "mutated"
+def test_backup_captures_state(live_app):
+    assert (
+        lifecycle.exec_in_app(live_app, ["cat", MARKER_PATH]).strip() == "original"
+    ), "the seeded state was not present at backup time"