ClusterMesh scale: Phase 3 — scale tiers + parallel CL2 fan-out

[skosuri1]

Stacked on top of #1157 (skosuri/clustermesh-scale). Do not merge until #1157 is merged; review/merge order matters.

This PR continues the ClusterMesh scale-test scenario with Phase 3 work — moving from harness validation (2 small clusters) to real scale measurement across cluster-count tiers.

Phase 3 Deliverables

• 20-node baseline cluster size (spec line 24). Current clusters are 3 nodes (2 default + 1 prompool) — sized for harness validation, not real scale measurement.
• Cluster-count tiers: add azure-5.tfvars, azure-10.tfvars, azure-20.tfvars and corresponding pipeline matrix entries. Each tier: validate quota, validate peering count (N·(N-1) at separate-VNet mode — 380 at N=20), tune CL2 timeouts, document breaking points.
• Parallel CL2 fan-out: replace sequential per-cluster CL2 with bounded concurrency (default 4). Requires async wrapping of utils.run_cl2_command (currently synchronous, modules/python/clusterloader2/utils.py:66-72) and confirming the AzDO agent has CPU/RAM headroom for N concurrent CL2 + Prometheus.
• etcd PodMonitor capacity check at 20 clusters: 28 watchers per cluster × 20 = 560 watchers; verify Prom scrape budget holds.
• Scaling-curve dashboards from cluster-attributed results (Kusto).

Out of Scope (deferred to later phases / pre-merge of #1157)

• Pre-merge housekeeping for Add Cilium ClusterMesh scale-test scenario #1157 (DEBUG-DUMP block, dev-pipeline placeholder revert, comment-trim pass) — stays on the base branch.
• Remaining scale scenarios (Allowing terraform inputs as JSON format #2 pod churn, add userdata bash scripts for lb eof error repro #3 node churn, Refactoring to add role concept #4 API server failure, Fix aws and azure issue #5 isolation, Change job id to run id #6 upper-bound, Refactor Terraform input variables #7 HA replicas) — Phase 4.