Add telemetry export of trace send rate, dropped traces, latency talking to refinery or real honeycomb

[lizthegrey]

Is your feature request related to a problem? Please describe.
It's hard to tell if we're failing to send requests from loadgen or they're going too slowly.

Describe the solution you'd like
Export otlp metrics to real honeycomb on how we're doing with our otlp send process with the loadtest SDK