Alternative for UserMetricsSystem into l4jv2 branch?

[mastermin]

We would like to upgrade our Databricks Runtime.
Currently we are on 10.4 and using UserMetricsSystem to create customer specific counter metrics.

Like this one:

import org.apache.spark.metrics.UserMetricsSystems

private final val metrics = UserMetricsSystems.getMetricSystem(
identifier,
builder => {
for (m <- metricTypes) {
builder.registerCounter(m.toString())
}
}
)

Branch l4jv2 doesn't contain the UserMetricsSystems class anymore.

README told me:
"
Databricks has contributed an updated version to support Azure Databricks Runtimes 11.0 (Spark 3.3.x) and above on the l4jv2 branch at: https://github.com/mspnp/spark-monitoring/tree/l4jv2.
Be sure to use the correct branch and version for your Databricks Runtime.
"

What I have to do to create/register/enable custom specific metrics (aka counter)?

Thanks for advice.

BR,
Markus