lab1 workers on both PVE nodes: per-node template + MD, balanced CA

PVE is now the 2-node cluster homelab (pv1-1 + pv1-2), no shared storage:
- template 101 copied to pv1-2 as 9101; MDs lab1-workers-pv1-1 (max 6) and
  lab1-workers-pv1-2 (max 12), PMTs pinned with allowedNodes; pool .31-.50;
  lab1-workers renamed for symmetry (decision #18, #7 superseded)
- CA: balancing-label node.cluster.x-k8s.io/pool (+ pool label, capacity labels
  with os/arch, maxPods on the MDs), scan-interval 1m, v 2, default QPS (#19).
  Scale test: 16 pods -> 4 + 4 across both nodes.
- runbook, image-builder copy procedure, fixes tables, future task to report
  three CA clusterapi bugs upstream, lab facts, changelog, execution log

Plan: docs/plans/2026-09-29-2315-pv1-2-workers.md

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

configs/cluster-autoscaler/values.yaml

@@ -18,7 +18,8 @@ rbac:
       verbs: [get, list, watch]
 extraArgs:
   expander: least-waste
-  # Aggressive lab scale-down profile (CA 1.35 defaults in brackets): idle worker gone in ~3 min.
+  # Aggressive lab scale-down profile (CA 1.35 defaults in brackets): idle worker gone in ~3-4 min
+  # (scan-interval 1m adds up to a minute to the 1m unneeded time).
   # Keep delay-after-add >= node startup (~2-3 min here) or bursts flap up/down.
   scale-down-unneeded-time: 1m                # [10m]
   scale-down-delay-after-add: 2m              # [10m] cluster-wide cooldown after any scale-up
@@ -27,7 +28,15 @@ extraArgs:
   scale-down-utilization-threshold: "0.6"     # [0.5]
   max-graceful-termination-sec: "60"          # [600] drain wait per pod
   balance-similar-node-groups: true
+  # One worker MD per PVE node. Compare groups only on this label: the generic comparator fails when
+  # one group has a real-node template (cached, 2Gi VM = ~1.78Gi allocatable, os/arch set) and the
+  # other a from-zero one. Disables the resource checks, so keep the MDs' capacity identical.
+  balancing-label: node.cluster.x-k8s.io/pool
+  # Fewer API requests instead of a higher kube-client-qps: every MD at 0 replicas costs ~3 uncached
+  # discovery GETs per NodeGroups() call (~15x per loop). Unschedulable pods and scale-ups/downs still
+  # trigger a loop immediately; only idle loops (scale-down evaluation) are spaced out.
+  scan-interval: 1m                           # [10s]
   # Cilium taints new nodes NoSchedule until its agent is up. Without this, CA simulates the freshly
   # registered (still tainted) node, finds the pending pods don't fit and adds one node too many.
   startup-taint: node.cilium.io/agent-not-ready
-  v: 4
+  v: 2                                        # [chart 4] 4 logs every throttled request and MD lookup