Why is it important to collect table statistics for Iceberg tables in watsonx.data?

Cost-based optimization relies on accurate statistics (row counts; column cardinality; null counts; histograms) to choose optimal join orders; aggregation strategies; and partition pruning; without them the optimizer makes poor choices. The incorrect options ("To determine which users have accessed the table in the past 30 days", "To automatically compact small Parquet files into larger ones", "To synchronize table metadata between the Hive Metastore and the object storage bucket") are distractors that don't fully capture the concept described.