Connectivity with Telegraf

## Pitch

While this metrics adapter makes JMX metrics consumable by Prometheus already, we are now looking at how they could be optimally consumed by Telegraf.

Opposed to our other integration with the Telegraf agent, which is [about using CrateDB as a metrics store](https://cratedb.com/docs/guide/integrate/telegraf/), this pitch is about **exposing CrateDB's metrics into monitoring environments running on Telegraf** instead of Prometheus, effectively feeding metrics data as a metrics _client_.

The issue names no files, tests, or entry points. Start by clarifying the Telegraf-facing integration and its expected metrics protocol; the issue does not specify completion criteria.

Written by the indexing model from the issue text.