Extend metrics wiring and demo/helm config

[novatechflow]

Summary

• Added broker-side runtime/lag/latency metrics and surfaced them in the console UI.
• Wired operator etcd snapshot metrics into the Advanced tab and demo/Helm defaults.
• Refined Ops UI layout (scalable broker list, 4‑column layout, slimmer metrics rail).

Key Changes

• Broker metrics: produce latency histogram, consumer lag (max + p95), runtime CPU/mem/uptime/goroutines.
• Console metrics: parses broker runtime metrics and operator snapshot metrics; exposes them in /ui/api/metrics.
• UI:
  • New broker fleet summary + collapsible list; removed misleading per‑broker partitions.
  • Live metrics moved to a right‑side rail with one card per row.
  • Advanced tab includes Admin metrics + ETCD snapshot status.
• Helm/demo: operator metrics URL auto‑wired via Helm; demo target uses operator metrics endpoint.

Notes

• Operator metrics are sourced from KAFSCALE_CONSOLE_OPERATOR_METRICS_URL (Helm now defaults this when operator metrics are enabled).
• Advanced tab shows n/a when operator metrics are unavailable.

Testing

• go test ./...
• make test-produce-consume (broker changes)
• make demo-platform (metrics)

Checklist

• Added/updated unit tests for new logic
• Added/updated e2e coverage for bug fixes
• Added license headers to new files