ci(vector-sync): relax single-user scan interval 5s->30s to stop churn flood

The single-user `mcp` container re-scanned the whole corpus every 5s. The
scanner already runs an initial scan on startup, so the corpus is indexed
without a frequent re-scan — the short interval only re-queues everything faster
than the 2 workers can drain. On a loaded CI runner pending_count snowballs
(observed 1465 pending, status stuck "syncing"), so freshly-created notes aren't
indexed within the tests' 90s sync-wait and test_rag / test_sampling cascade
into 300s pytest timeouts (intermittent single-user failures; passed on nc33 /
faster runs).

Raise it to 30s, matching the already-tuned multi-user-basic cadence and staying
well within the 90s wait budget. multi-user-basic (30s) and login-flow (60s) are
already relaxed; this brings single-user in line.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

docker-compose.yml

@@ -99,7 +99,15 @@ services:
 
       # Semantic search configuration (ADR-007, ADR-021)
       - ENABLE_SEMANTIC_SEARCH=true
-      - VECTOR_SYNC_SCAN_INTERVAL=5
+      # The sync scanner runs an initial scan immediately on startup, so the
+      # corpus is indexed without waiting for the first periodic re-scan. A very
+      # short re-scan interval (was 5s) just re-queues the whole corpus faster
+      # than the workers can drain it: on a loaded CI runner pending_count
+      # snowballs (observed 1465 pending), sync never reaches idle, and the
+      # single-user vector tests (test_rag, test_sampling) time out. 30s matches
+      # the tuned multi-user-basic cadence and stays well within the tests' 90s
+      # sync-wait budget.
+      - VECTOR_SYNC_SCAN_INTERVAL=30
       - VECTOR_SYNC_PROCESSOR_WORKERS=2
       # Required to enable the /webhooks/nextcloud receiver (GHSA-8vh3-g2qg-2h2c).
       # Without it the route is not mounted and vector sync relies on polling.