Better support for histogram metrics

[Kasia Kujawa]

Now to extract _sum and count for histogram metics it is need to make following changes in configuration, for example for jvm_gc_collection_seconds metric

sumologic:
  metrics:
    collector:
      otelcol:
        config:
          merge:
            processors:
              transform/extract_sum_count_from_histograms:
                error_mode: ignore
                metric_statements:
                - context: metric
                  statements:
                  - extract_sum_metric(true) where IsMatch(name, "^(jvm_gc_collection_seconds|apiserver_request_duration_seconds|coredns_dns_request_duration_seconds|kubelet_runtime_operations_duration_seconds)$")
                  - extract_count_metric(true) where IsMatch(name, "^(jvm_gc_collection_seconds|apiserver_request_duration_seconds|coredns_dns_request_duration_seconds|kubelet_runtime_operations_duration_seconds)$")

metadata:                   
  metrics:
    config:
      merge:
        processors:
          filter/drop_unnecessary_metrics:
            error_mode: ignore
            metrics:
              metric:
              - resource.attributes["job"] != "pod-annotations" and IsMatch(name, "scrape_.*")
              - IsMatch(name, "^(jvm_gc_collection_seconds|apiserver_request_duration_seconds|coredns_dns_request_duration_seconds|kubelet_runtime_operations_duration_seconds)$")

Related pull request in which this basic support for histogram metrics was added: #3192

I would be good to have easier mechanism to extract _sum and _count` for example by providing list of regular expressions for histogram metrics in the values.yaml