Fix incorrect NDV (1) for VARCHAR columns in DBA_TAB_COL_STATISTICS during online statistics gathering with online_osg_sample / ONLINE_ESTIMATE_PERCENT

Description

During online statistics gathering (specifically in the online_osg_sample test case related to the ONLINE_ESTIMATE_PERCENT preference), the NDV (Number of Distinct Values) for VARCHAR columns in the DBA_TAB_COL_STATISTICS view is incorrectly reported as 1.

Actual Behavior:
The NDV for VARCHAR columns is 1.

Expected Behavior:
The NDV should reflect the actual number of distinct values in the column.

Steps to Reproduce

1. Log into the test machine.

2. Navigate to the directory: /obdata/data/AndroidTest/ob_opensource_test/tools/deploy.

3. Execute the following commands to set up the test environment and run the specific mysqltest case:

   export OCEANBASE_HOME=/obdata/data/AndroidTest/ob_opensource_test/
   cd ${OCEANBASE_HOME}/tools/deploy
   source activate_obd.sh
   export OBD_SEEKDB_STATUS_FALLBACK_TCP=1
   
   obd test mysqltest test_env --mysqltest-bin=${OCEANBASE_HOME}/deps/3rd/u01/obclient/bin/mysqltest --obclient-bin=${OCEANBASE_HOME}/deps/3rd/u01/obclient/bin/obclient --init-only --init-sql-dir=${OCEANBASE_HOME}/tools/deploy --init-sql-files='init.sql,init_user.sql|root@sys|test'
   obd test mysqltest test_env --mysqltest-bin=${OCEANBASE_HOME}/deps/3rd/u01/obclient/bin/mysqltest --obclient-bin=${OCEANBASE_HOME}/deps/3rd/u01/obclient/bin/obclient --disable-reboot --test-set=online_opt_stat_gather.online_osg_sample
   

4. The test case executes the following SQL sequence which reproduces the issue:

   USE test;
   
   DROP TABLE IF EXISTS t1, t2, t3, t4, t5;
   DROP PROCEDURE IF EXISTS batch_insert;
   
   CREATE TABLE t1 (c1 BIGINT PRIMARY KEY, c2 VARCHAR(256), c3 VARCHAR(256), c4 DATETIME, c5 DATETIME)
     WITH COLUMN GROUP (each column);
   CREATE TABLE t2 LIKE t1;
   CREATE TABLE t3 LIKE t1;
   CREATE TABLE t4 LIKE t1;
   CREATE TABLE t5 LIKE t1;
   
   DELIMITER $$
   CREATE PROCEDURE batch_insert(IN args INT)
   BEGIN
     DECLARE i INT;
     SET i = 0;
     START TRANSACTION;
     WHILE i < args DO
       INSERT INTO t1(c1,c2,c3,c4,c5) VALUES (
         i,
         CONCAT('C2_MKKHJHHS_XXXSDSDSDDSDSSJJJJDSHH_', i),
         CONCAT('C3_MKKHJHHS_XXXMMMMMMJJJDSHH_', i),
         '2023-01-01 01:11:11',
         '2023-01-01 01:11:11'
       );
       SET i = i + 1;
       IF i % 10000 = 0 THEN COMMIT; END IF;
     END WHILE;
     COMMIT;
   END$$
   DELIMITER ;
   
   CALL batch_insert(100000);
   
   CALL dbms_stats.set_global_prefs('ONLINE_ESTIMATE_PERCENT', '1');
   INSERT /*+ query_timeout(100000000) append parallel(2) enable_parallel_dml */ INTO t2 SELECT c1, c2, c3, c4, c5 FROM t1;
   INSERT /*+ query_timeout(100000000) append parallel(2) enable_parallel_dml */ INTO t3 SELECT c1, c2, c3, c4, c5 FROM t1;
   
   CALL dbms_stats.set_global_prefs('ONLINE_ESTIMATE_PERCENT', '50');
   INSERT /*+ query_timeout(100000000) append parallel(2) enable_parallel_dml */ INTO t4 SELECT c1, c2, c3, c4, c5 FROM t1;
   INSERT /*+ query_timeout(100000000) append parallel(2) enable_parallel_dml */ INTO t5 SELECT c1, c2, c3, c4, c5 FROM t1;
   
   CALL dbms_stats.set_global_prefs('ONLINE_ESTIMATE_PERCENT', '1');
   
   SELECT TABLE_NAME, NUM_ROWS, SAMPLE_SIZE
   FROM oceanbase.DBA_TAB_STATISTICS
   WHERE table_name IN ('t2','t3','t4','t5')
   ORDER BY table_name;
   
   SELECT TABLE_NAME, COLUMN_NAME, NUM_DISTINCT, LOW_VALUE, HIGH_VALUE
   FROM oceanbase.DBA_TAB_COL_STATISTICS
   WHERE table_name IN ('t2','t3','t4','t5')
   ORDER BY table_name, column_name;
   

The final queries will show that the NUM_DISTINCT for the VARCHAR columns (c2, c3) is 1, which is incorrect.

Start by running the online_opt_stat_gather.online_osg_sample mysqltest case with the supplied setup commands, then inspect the online statistics gathering implementation for VARCHAR NDV handling. Confirm the fix by querying oceanbase.DBA_TAB_COL_STATISTICS for c2 and c3 in t2 through t5 and verifying NUM_DISTINCT reflects the inserted distinct values.