NEO-130: Address review feedback on unlock_granted stub.

Document E9.M1 durable-emit gating for rollback paths; reconcile plan
bullet for class summary; strike through resolved review items.

docs/plans/NEO-130-implementation-plan.md

@@ -50,7 +50,7 @@
 
 ## Implementation reconciliation (shipped)
 
-- **Engine anchor:** **`RewardRouterOperations.TryDeliverQuestCompletion`** — two comment-only hook sites with **`TODO(E9.M1)`** (no runtime emit).
+- **Engine anchor:** **`RewardRouterOperations.TryDeliverQuestCompletion`** — two comment-only hook sites with **`TODO(E9.M1)`** (no runtime emit); class xml summary cites NEO-130 telemetry anchor.
 - **`reward_delivery`:** after **`deliveryStore.TryRecord`** returns **`true`** on first-time delivery.
 - **`unlock_granted`:** stub after item + skill XP apply loops, before delivery event build — future **`UnlockGrant`** apply region; prototype bundles have no unlock rows.
 - **Docs:** `server/README.md` reward telemetry subsection; E7.M2 module + alignment register + E7M2 backlog updated.

docs/reviews/2026-06-07-NEO-130.md

@@ -2,7 +2,8 @@
 
 **Date:** 2026-06-07  
 **Scope:** Branch `NEO-130-reward-telemetry-hooks` vs `origin/main` — commits `7493a6b` … `c461a82`  
-**Base:** `origin/main`
+**Base:** `origin/main`  
+**Follow-up:** Review suggestions/nits addressed in a subsequent commit on the same branch.
 
 ## Verdict
 
@@ -33,13 +34,13 @@ NEO-130 adds **comment-only** E9.M1 telemetry hook sites to **`RewardRouterOpera
 
 ## Suggestions
 
-1. **E9.M1 `unlock_granted` wiring note** — The stub sits before **`TryRecord`**, so a future literal emit at this line would also run on the race-rollback path (grants applied then reverted). When **`UnlockGrant`** rows land, place per-row emits inside the apply loop and/or gate on durable commit (mirror **`reward_delivery`** after **`TryRecord`** success). The stub comment already says “once per applied UnlockGrant row”; consider adding one line that durable emit should not fire on rollback paths when E9.M1 implements the loop.
+1. ~~**E9.M1 `unlock_granted` wiring note** — The stub sits before **`TryRecord`**, so a future literal emit at this line would also run on the race-rollback path (grants applied then reverted). When **`UnlockGrant`** rows land, place per-row emits inside the apply loop and/or gate on durable commit (mirror **`reward_delivery`** after **`TryRecord`** success). The stub comment already says “once per applied UnlockGrant row”; consider adding one line that durable emit should not fire on rollback paths when E9.M1 implements the loop.~~ **Done.** Stub comment now gates E9.M1 durable emit to apply-loop and/or post-**`TryRecord`** paths only (not race rollback).
 
 ## Nits
 
-- Nit: **`unlock_granted`** stub runs even for empty bundles (no item/skill rows) — correct for a region marker; E9.M1 loop will simply emit nothing until unlock rows exist in content.
+- ~~Nit: **`unlock_granted`** stub runs even for empty bundles (no item/skill rows) — correct for a region marker; E9.M1 loop will simply emit nothing until unlock rows exist in content.~~ **Done.** Acknowledged; no change.
 
-- Nit: Plan technical approach lists class-summary update on **`RewardRouterOperations`**; shipped in xml summary — reconciliation bullet could mention it for parity with other stories (cosmetic doc only).
+- ~~Nit: Plan technical approach lists class-summary update on **`RewardRouterOperations`**; shipped in xml summary — reconciliation bullet could mention it for parity with other stories (cosmetic doc only).~~ **Done.** Reconciliation bullet updated.
 
 ## Verification
 

server/NeonSprawl.Server/Game/Rewards/RewardRouterOperations.cs

@@ -140,6 +140,8 @@ public static class RewardRouterOperations
         // Prototype completionRewardBundle has item + skill XP rows only; no runtime unlock apply in Slice 2.
         // Planned payload fields: playerId, questId, unlock id / kind — finalize at E9.M1 wiring time.
         // Distinct from `perk_unlock` (PerkUnlockEngine, NEO-49) which may fire as a side effect of skill XP grants.
+        // E9.M1 wiring: place per-row emits inside the apply loop and/or after TryRecord success only — not on
+        // TryRecord race-rollback paths (grants applied then reverted below); mirror reward_delivery gating.
 
         var deliveredAt = timeProvider.GetUtcNow();
         var deliveryEvent = new RewardDeliveryEvent(