CDC stats | Latency, average pipeline frequency, backpressure signals

As a user, i want to lookup the current latency similar to mariadb replication (Seconds_Behind_Master) to know if my CDC is up to date or delayed by x milliseconds.
Furthermore what the average time to flush a pipeline/micro batch into exasol.
As well as any indication if the lag is growing or shrinking via a backpressure kpi to understand how far behind we are, maybe how much space is being used because we're behind etc. anything relevant a user would want to know or needs to monitor if CDC cant keep up.

This information will be used by enterprise manager and mariadb cloud monitoring.

• Latency — time from source event commit to downstream apply (end-to-end lag), plus per-stage breakdown if feasible (capture, transform, apply).
• Average pipeline frequency — rate of events/batches processed over a rolling window, to establish a baseline throughput.
• rows over the last 60 seconds versus mb or kb of data over the last 60s
• Backpressure signals — indicator(s) when the pipeline is being throttled or queuing due to downstream overload saturation (e.g., queue depth, buffer utilization, or explicit backpressure flag).
• max RAM over the last hour?