Broker routing time is impact by the time pruner with 10k+ segments

We seen for tables where we have 10 thousand or more partitions, broker routing time will increase to 100ms or more. We have not had a chance to profile yet, but the theory is that serializing/deserializing interval trees with large numbers of segments is causing issues.

Start by profiling broker routing for tables with 10,000 or more partitions and inspect the time pruner's interval-tree serialization and deserialization path. Confirm whether those operations account for the reported 100ms or higher routing time, then validate the result against a large-segment case.

Written by the indexing model from the issue text.