Add Prometheus metrics endpoint for monitoring and observability

[treblereel]

Context

Part of epic #1 — Build production-ready REST microservice with k8s deployment on top of casehub-engine. Production deployments need metrics for monitoring service health, performance, and business KPIs.

What

Configure Prometheus metrics endpoint using Micrometer:

• /q/metrics — Prometheus-format metrics endpoint
• Standard JVM metrics (memory, GC, threads)
• HTTP request metrics (request count, duration, status codes)
• Custom case engine metrics (active cases, completed cases, faulted cases, worker executions)

Acceptance Criteria

• Prometheus metrics endpoint available at /q/metrics
• JVM metrics exposed (heap usage, GC pauses, thread count)
• HTTP metrics exposed (request rate, latency percentiles, error rate)
• Custom gauge: casehub_cases_active — count of currently running cases
• Custom counter: casehub_cases_completed_total — total completed cases
• Custom counter: casehub_cases_faulted_total — total faulted cases
• Custom counter: casehub_worker_executions_total — total worker executions (with labels: worker name, status)
• Integration tests verify metrics endpoint returns valid Prometheus format

Notes

• Use quarkus-micrometer-registry-prometheus extension
• Consider adding histogram for case duration: casehub_case_duration_seconds
• Metrics should update in real-time as cases execute
• Labels for custom metrics: namespace, case name, version
• File paths: metrics producer classes, MeterRegistry injection points