Expose Prometheus metrics for monitoring cardano-rosetta-java API

[linconvidal]

Why

Monitoring the performance and health of the cardano-rosetta-java API is essential for maintaining high availability and reliability. Currently, cardano-node already supports Prometheus metrics for monitoring purposes, providing insights into node health, transaction processing, and resource usage. However, the cardano-rosetta-java API itself does not expose Prometheus metrics.

What

TBD (wip)

Metrics could include (but are not limited to):

• Request Metrics:
  • Total number of requests per endpoint?
  • Request duration (latency) per endpoint?
  • HTTP response codes frequency?
• Application Metrics:
  • JVM memory usage (heap/non-heap)?
  • JVM garbage collection statistics?
  • Thread pool utilization?
• Error Metrics:
  • Exception occurrences by type?
  • Failure rate per endpoint?

Prometheus metrics endpoint (is that okay according to Mesh standards?):

GET /metrics