Implement queue metrics for the disk queue

[fearful-symmetry]

The disk queue currently does not implement a metrics API:

https://github.com/elastic/beats/blob/d04da481f51c3f41b814106259145042baace857/libbeat/publisher/queue/diskqueue/queue.go#L284-L286

These are metrics used by self-monitoring in the shipper, as part of this issue: #11

These metrics API was designed with the memory queue in mind, so I'm not sure to what degree these metrics make sense for the disk queue, or how hard they might be to implement. Also not sure if this is something we want done by 8.6.

CC @cmacknz

[cmacknz]

The metrics to report are defined in #11, specifically we want:

• The current configured maximum size of the disk queue in bytes.
• The current utilization of the disk queue in bytes:
• The age of the oldest item in the disk queue in nanoseconds.

[pierrehilbert]

@fearful-symmetry we need to check if elastic/beats#33471 is closing this one or not