fix: remove bucket-init job, wait for pods before readiness check

- Remove minio-bucket-init Job entirely (seed_minio.py creates bucket)
- Wait for pods to exist before kubectl wait --for=condition=ready
- Fixes 'no matching resources found' race when pods are still ContainerCreating

infra/inttest/run_pipeline.sh

@@ -312,14 +312,6 @@ if ! kubectl wait --for=condition=ready pod -l app=minio -n "$NAMESPACE" --timeo
   exit 2
 fi
 
-log "Waiting for minio-bucket-init job ..."
-if ! kubectl wait --for=condition=complete job/minio-bucket-init -n "$NAMESPACE" --timeout=120s; then
-  log "WARNING: minio-bucket-init job did not complete within 120s"
-  log "Bucket-init pod logs:"
-  kubectl logs -l app=minio-bucket-init -n "$NAMESPACE" --tail=30 2>&1 || true
-  kubectl describe job/minio-bucket-init -n "$NAMESPACE" 2>&1 | tail -20 || true
-fi
-
 stage_end "infra_deploy" "ok"
 
 # ══════════════════════════════════════════════════════════════════════════════
@@ -402,6 +394,15 @@ envsubst < "$REPO_ROOT/infra/inttest/services.yaml" \
   | kubectl apply -n "$NAMESPACE" -f -
 
 log "Waiting for all API services to become ready ..."
+# Wait for pods to be created by the ReplicaSet controller before checking readiness.
+# kubectl wait fails immediately with "no matching resources found" if no pods exist yet.
+for i in $(seq 1 30); do
+  POD_COUNT=$(kubectl get pods -n "$NAMESPACE" -l tier=api --no-headers 2>/dev/null | wc -l)
+  if [ "$POD_COUNT" -ge 4 ]; then
+    break
+  fi
+  sleep 2
+done
 if ! kubectl wait --for=condition=ready pod -l tier=api -n "$NAMESPACE" --timeout=120s; then
   log "FATAL: API services did not become ready"
   log "Pod statuses:"