[Bug]: load data test performance down

[heni02]

Is there an existing issue for the same bug?

• I have checked the existing issues.

Branch Name

main

Commit ID

fd9d868

Other Environment Information

- Hardware parameters:
- OS type:
- Others:

Actual Behavior

load data 性能数据：https://grafana.ci.matrixorigin.cn/d/U7ZX6G6Vk/test-nightly-tke-loaddata?orgId=1，main分支有性能下降的现象，
tpch1T load时由之前1个小时耗时增长到2个小时+

二分结果：

Expected Behavior

No response

Steps to Reproduce

create table table_20_columns(
clo1 tinyint,
clo2 smallint,
clo3 int,
clo4 bigint,
clo5 tinyint unsigned,
clo6 smallint unsigned,
clo7 int unsigned,
clo8 bigint unsigned,
col9 float,
col10 double,
col11 varchar(255),
col12 Date,
col13 DateTime,
col14 timestamp,
col15 bool,
col16 decimal(5,2),
col17 text,
col18 varchar(255),
col19 varchar(255),
col20 varchar(255)
);

load data url s3option {'endpoint'='http://cos.ap-guangzhou.myqcloud.com','access_key_id'='xxxx','secret_access_key'='xxx','bucket'='mo-load-guangzhou-1308875761', 'filepath'='100_columns/100000000_100_columns_load_data.csv', 'compression'='none'} into table test.table_100_columns fields terminated by ',' lines terminated by '\n' parallel 'true';

Additional information

No response

[heni02]

下降commit时的profile：
https://grafana.ci.matrixorigin.cn/explore?panes=%7B%22WTr%22:%7B%22datasource%22:%22pyroscope%22,%22queries%22:%5B%7B%22groupBy%22:%5B%5D,%22labelSelector%22:%22%7Bnamespace%3D%5C%22mo-main-nightly-fd9d86896-20250120%5C%22%7D%22,%22queryType%22:%22both%22,%22refId%22:%22A%22,%22profileTypeId%22:%22memory:alloc_objects:count:space:bytes%22,%22datasource%22:%7B%22type%22:%22grafana-pyroscope-datasource%22,%22uid%22:%22pyroscope%22%7D%7D%5D,%22range%22:%7B%22from%22:%221737390840000%22,%22to%22:%221737390938000%22%7D%7D%7D&schemaVersion=1&orgId=1

[ouyuanning]

暂定不回退。会先调整一下刷s3策略，找一个平衡点吧。

如果完全回退，容易碰到很少数据也刷s3，导致小s3文件过多，后面合并压力过大的问题。