[ops dashboard] Partition Balance needs a rename

          "expr": "100 * abs(1-(sum(stddev by (redpanda_topic) (sum(redpanda_kafka_max_offset{redpanda_namespace=\"kafka\",redpanda_cloud_data_cluster_name=~\"\"}) by (redpanda_topic,redpanda_partition))) /  sum(avg by (redpanda_topic) ((sum(redpanda_kafka_max_offset{redpanda_namespace=\"kafka\",redpanda_cloud_data_cluster_name=~\"\"}) by (redpanda_topic,redpanda_partition))))))",

This currently shows the balance of writes to partitions across the cluster. It's confusingly named since we also think about the balance of partition replicas (data) and partition leadership (which brokers own updates to a primary partition and replication to the partition followers).

I get the intent of this to show how well the data is distributed on writes, but I'm not sure we're being clear on this. We probably need to rename this to something like "Partition Write Distribution" and give it an info block so people understand how to interpret.

On a cluster with only a handful of topics, it's probably ok to interpret this as topic write evenness, but on complex clusters with hundreds or thousands of topics/partitions, I'm not sure how this should really come across because the write loads could be fairly skewed to a subset of topics and still have a relatively even write load.

Start from the dashboard panel containing the provided PromQL expression and inspect how its current title and explanatory text are defined. Confirm the intended distinction between write distribution, replica balance, and leadership, then rename the panel and add an info block that explains its interpretation; done means the dashboard wording matches that distinction.