ws-worker: tracing events

Most of what the worker does is listen to events going on inside the engine and reporting them back to lighting. As well as managing the work loop to pull attempts from lightining.

But the worker is going to get an extra responsibility: tracing.

We want to be able to attach an event listener to the worker to track what's going on inside it. We should listen to workflow and job star/tcomplete and stuff.

But we also want to monitor the loading of dataclips, the time it takes to send messages to lightning (especially messages with dataclips attached), the loading of credentials, errors which occur, and so on.

These events can be pushed out to prometheus or whatever to track metrics and general behaviour.

I don't think it's a particularly big job, there's already an event hook on the worker. It just needs rationalising a bit - like should it directly emit events from the inner worker or catch and wrap them? Also monitoring of lightning comms needs adding seperately.

Start at the existing event hook on the worker and trace how events from the inner worker, workflow and job lifecycle, dataclip loading, credentials, errors, and Lightning communication flow through it. Clarify the event surface and wrapping strategy with maintainers, then verify that the agreed tracing events expose the listed worker activity for metrics consumers.