feat: implement prometheus metrics for base node

[sdbondi]

Description

• Instrument tari_comms for prometheus metrics
• add metrics feature to base node and tari_comms
• add prometheus metrics scrape server
• add prometheus push client
• add configuration for metrics common.metrics
• add get-network-stats command
• fix bug in value for last request latency

Motivation and Context

Increase base node visibility

Dependencies

• warp for http metric pull server
• reqwest for metric push client
• prometheus for instrumentation

How Has This Been Tested?

Manually

• run prometheus docker run --rm -it -p 9090:9090 -v `pwd`/prometheus.yml:/etc/prometheus/prometheus.yml prom/prometheus
• run base node(s) with --metrics-bind-addr 127.0.0.1:5544
• visit localhost:9090

# prometheus.yml
scrape_configs:
  - job_name: 'node1'
    scrape_interval: 5s
    metrics_path: "/metrics"
    static_configs:
      - targets: ['host.docker.internal:5544']
        labels:
          node: 'node1'

[delta1]

Looks good will try it out. Random test failure that might have been fixed on dev?

[stringhandler]

Nice. I'm personally not a big fan of having many different compile time features, because you then have to create a matrix of possibilities to compile and test against. But that said, I see why you did it here, and I might have done the same.

Happy for it to go in