load hive data into openmldb error: requirement failed: schema mismatch

**Bug Description**
openMLDB:0.9.2
load hive data into openmldb error: requirement failed: schema mismatch

**Expected Behavior**

**Relation Case**

**Steps to Reproduce**

1. hive table : CREATE EXTERNAL TABLE `tmp.ods_admin_da`(
`id` bigint,
`biz_key` string ',
`biz_domain` string ,
`biz_type` string ,
`locked` string ,
`url` string ,
`method` string ,
`created_at` string,
`updated_at` string,
`biz_tag` string,
`status` string)
2. openMLDB table: desc test4;
---- ------------ --------- ------ ---------
# Field Type Null Default
---- ------------ --------- ------ ---------
1 id BigInt YES
2 biz_key Varchar YES
3 biz_domain Varchar YES
4 biz_type Varchar YES
5 locked Varchar YES
6 url Varchar YES
7 method Varchar YES
8 created_at Varchar YES
9 updated_at Varchar YES
10 biz_tag Varchar YES
11 status Varchar YES
3. LOAD DATA INFILE 'hive://tmp.ods_admin_da' INTO TABLE demo_db.test4 OPTIONS(deep_copy=true,mode='append');
4. Get exception: requirement failed: schema mismatch, loaded StructType(StructField(id,LongType,true),StructField(biz_key,StringType,true),StructField(biz_domain,StringType,true),StructField(biz_type,StringType,true),StructField(locked,StringType,true),StructField(url,StringType,true),StructField(method,StringType,true),StructField(created_at,StringType,true),StructField(updated_at,StringType,true),StructField(biz_tag,StringType,true),StructField(status,StringType,true)) != table StructType(StructField(id,LongType,true),StructField(biz_key,StringType,true),StructField(biz_domain,StringType,true),StructField(biz_type,StringType,true),StructField(locked,StringType,true),StructField(url,StringType,true),StructField(method,StringType,true),StructField(created_at,StringType,true),StructField(updated_at,StringType,true),StructField(biz_tag,StringType,true),StructField(status,StringType,true)), check hive://kc_tmp.ods_admin_admin_biz_config_da
5.

The error indicates a schema mismatch between a Hive table and an OpenMLDB table, despite identical field names and types. Look at the data loading code, likely in the Spark connector or Hive integration module. Check how schemas are compared and why they are considered mismatched. Run the provided LOAD DATA command in a test environment to reproduce the issue.

Written by the indexing model from the issue text.