Let's add autoscaling support for our helm chart, with the following configuration:
marquez.autoscaling.enabled - enable autoscaling
marquez.autoscaling.minReplicas - minimum number of replicas
marquez.autoscaling.maxReplicas - maximum number of replicas
marquez.autoscaling.targetCPU - target CPU utilization percentage
marquez.autoscaling.targetMemory - target Memory utilization percentage