server.js: /api/health now includes disk usage (df -B1 /) and load_avg
(os.loadavg) + cpu_count. station.html: disk bar, load average chips
with warn/crit coloring vs cpu count, alert banner highlights issues.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Single snapshot returns lifetime average (near 0 on idle systems).
Two samples 100ms apart gives real-time cpu usage per core, then averaged.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>