[docs] Update queries of "Column Pruning" section to align with DDL column definitions (#1190)

Author: ZmmBigdata

website/docs/table-design/table-types/log-table.md

@@ -32,7 +32,8 @@ CREATE TABLE log_table (
   order_id BIGINT,
   item_id BIGINT,
   amount INT,
-  address STRING
+  address STRING,
+  dt DATE
 )
 WITH ('bucket.num' = '3');
 ```
@@ -73,10 +74,10 @@ During query execution, query engines like Flink analyzes the query to identify
 For example the following streaming query:
 
 ```sql
-SELECT id, name FROM log_table WHERE timestamp > '2023-01-01';
+SELECT order_id, item_id FROM log_table WHERE dt > '2023-01-01';
 ```
 
-In this query, only the `id`, `name`, and `timestamp` columns are accessed. Other columns (e.g., `address`, `status`) are pruned and not read from storage.
+In this query, only the `order_id`, `item_id`, and `dt` columns are accessed. Other columns (e.g., `address`, `amount`) are pruned and not read from storage.
 
 
 ## Log Compression