Use context for metrics on how long it takes a message to get through pipeline

[maisiesadler]

Equivalent to this https://github.com/justeat/JustSaying/blob/322a3439aff8a649f882e7058a7b26808387ba72/src/JustSaying/Messaging/MessageProcessingStrategies/Throttled.cs#L115

We have this extension https://github.com/justeat/JustSaying/blob/1c3f53c3152141704cf76a97e95956bc5b3ec455/src/JustSaying/Messaging/Monitoring/MonitorExtensions.cs#L5

which is called here https://github.com/gkinsman/JustSaying/blob/a8f5a2d0c9988f58a137f536abe85a63d0ba37f0/src/JustSaying/Messaging/Channels/Receive/MessageReceiveBuffer.cs#L78

But it's not quite the same, because the previous one is a message has timed out and been chucked, whereas ours identifies any messages that have been held waiting to get into the multiplexer

We discussed that it might be good to use the context to send metrics for various parts of the pipeline, to track each message

[stale]

This issue has been automatically marked as stale because it has not had any recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[stale]

Closed automatically due to inactivity.

[stale]

This issue has been automatically marked as stale because it has not had any recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[gkinsman]

I think this better addressed by the OpenTelemetry work in #874. Will close for now, but please reopen if you disagree!