Change Prometheus Exporter Categorizations

[ralphm]

No description provided.

[ralphm]

A few remarks:

• Why is the GitHub collector in apm, but the GitHub API limit one in generic-data-collection?
• Is generic-data-collection meant as a catch-all, or specifically for generic collection like StatsD, SNMP, Prometheus?
• Maybe Cilium Hubble belongs in `networking-stack-and-network-interfaces'
• MQTT Blackbox: it is now in synthetic-checks, but maybe message-brokers is better?
• What is the difference between the APM and Observability categories?
• Why do we have a separate category for Kubernetes, but lump all other container and VM exporters into containers-and-vms?
• SSH: unsure, but authentication-and-authorization seems off
• What is the notifications category? It has StatusPage and OTRS. What about Discourse?
• bfptrace -> linux-systems?
• AMD CPU & GPU: linux-systems?
• Zulip is in notifications but XMPP is in message broker. Both are chat systems. Maybe we can lump them together into a Chat category?

[ilyam8]

cc @shyamvalsan (you added initial categories in #1253)

[shyamvalsan]

@ralphm @ilyam8 The base categories were just using what existed already on the website. And we need to fix stuff for sure.

• Generic data collection is intended to cover prometheus, statsd, snmp scenarios, where it can be used to collect data from a variety of sources.
• GitHub and GitHub API should both be somewhere else
• If we need a catch all, it should be "Other"
• I think @sashwathn already moved MQTT blackbox to message-brokers, PR pending
• Obervability is just about monitoring observability services
• Kubernetes vs Containers & VMs (I think the ask for a separate category for Kubernetes is from @ktsaou, we're just following the precedence from the website here)
• The notifications category has been renamed by @sashwathn to "Notifications and Incident Management" - not sure Discourse fits in either way