Thanks for sharing this. I'm afraid we're not accepting new modules, so I'd suggest maintaining it in your own repo. We can add a link to it from the docs.

I would suggest a read of https://prometheus.io/docs/instrumenting/writing_exporters/ in order to improve your metrics. In particular counters and only counters should end in _total, prefer exposing raw numerators/denominators rather than ratios, metrics should have units, and don't expose per-second values - rather expose the counter and let it be calculated in Prometheus. The latency_avg are unfortunately as the library isn't providing great stats, but I'd drop the "avg" (it's implied) and mention in the help over what time frame it is calculated.