Xinference last model load duration gauge

A GAUGE holding the duration of the LAST model load, not a histogram and not a counter. There is no distribution of load times on this engine - only the most recent one - so a slow-load regression that has since been overwritten by a fast load is invisible. The name ends _duration_seconds but the type is gauge and no bucket ladder exists. Xinference main (xorbitsai/inference), read 2026-10-05. NOT ON BY DEFAULT on the worker: the /metrics endpoint is only mounted when XINFERENCE_DISABLE_METRICS is unset AND a metrics exporter host or port is supplied; otherwise launch_metrics_export_server() is never called and the endpoint does not exist. Supervisor-side metrics are deregistered from the worker's registry on purpose (launch_metrics_export_server removes _SUPERVISOR_ONLY_METRICS), so the worker endpoint carries only worker metrics and the supervisor endpoint only supervisor metrics. aioprometheus, not prometheus_client: counters do not auto-suffix _total the way prometheus_client does, so two of these names end in _counter and three do not - match on the literal name.