Add monitoring with Prometheus metrics

Set up Prometheus metrics to help monitor edge-proxy health.

- [ ] Per-URL metrics similar to existing in core API, e.g. read/write ops, throughput, error rates
- [ ] Cache-related metrics, e.g. hit/miss

> [!IMPORTANT]
> Moving forwards, if we're to adopt the Rust implementation, this issue may be moved, or the implementation ported.