[inference] Add FTR coverage for the inference adapter/connector

[pgayvallet]

Follow-up of #204541

Now that we're supporting the inference connector, we should add FTR coverage for it, similar to what we've been doing for the other inference adapters.

This is going to be slightly more tricky, because inference connectors can't be "preconfigured", so we will have to create the stack connector from the UI or using the API (or eventually preconfiguring the connector and then just creating the inference endpoint)

[elasticmachine]

Pinging @elastic/appex-ai-infra (Team:AI Infra)