rpk: add load-factor dashboard to generate grafana-dashboard

Add the Redpanda Load Factor dashboard (merged to redpanda-data/observability
in PR #51) as a new --dashboard option. The dashboard shows utilization
relative to capacity for key resources (CPU, IO scheduler, disk IOPS, memory,
network bandwidth, client connections) using public metrics, to help identify
which resource a cluster will exhaust first.

Like the other dashboards, it is fetched from the observability repo at
runtime, with an embedded gzipped copy as the offline fallback.