hive sql查询hudi分区表，如果分区字段不是表最后一列，解析parquet文件后返回的数据，没有查询分区字段单在分区字段列位置自动增加了分区字段的值，导致后续列错误发生类型转换问题

**_Tips before filing an issue_**

- Have you gone through our [FAQs](https://hudi.apache.org/learn/faq/)?

- Join the mailing list to engage in conversations and get faster support at dev-subscribe@hudi.apache.org.

- If you have triaged this as a bug, then file an [issue](https://issues.apache.org/jira/projects/HUDI/issues) directly.

**Describe the problem you faced**

A clear and concise description of the problem.

**To Reproduce**

Steps to reproduce the behavior:

1.flink sql create table and insert data，use "fqsj" for the partition column ，and this column is not the last column
CREATE CATALOG hoodie_catalog
WITH (
'type'='hudi',
'catalog.path' = '/tmp/hudi',
'hive.conf.dir' = '/opt/hive-2.3.9/conf',
'mode'='hms'
);
USE CATALOG hoodie_catalog;
create database hoodie_catalog.flink_hudi;
CREATE CATALOG myhive WITH (
'type' = 'hive',
'hive-conf-dir' = '/opt/hive-2.3.9/conf'
);
CREATE TABLE hoodie_catalog.flink_hudi.TEST_COW(
jllsh VARCHAR(100),
syh VARCHAR(50) PRIMARY KEY NOT ENFORCED,
jzlsh VARCHAR(40),
grbsh VARCHAR(20),
dalx VARCHAR(10),
dzjkkkh VARCHAR(40),
yzzh VARCHAR(40),
yzbs VARCHAR(40),
fqsj VARCHAR(10),
dj DOUBLE,
ze DOUBLE,
zzsj TIMESTAMP
)
PARTITIONED BY (`fqsj`)
WITH (
'connector' = 'hudi',
'path' = 'hdfs://bigdata01:8020/tmp/hudi/flink_hudi/TEST_COW',
'table.type' = 'COPY_ON_WRITE',
'precombine.field' = 'syh',
'write.operation' = 'upsert',
'hoodie.datasource.hive_sync.support_timestamp' = 'true'
);
insert into hoodie_catalog.flink_hudi.TEST_COW select jllsh,syh,jzlsh,grbsh,dalx,dzjkkkh,yzzh,yzbs,fqsj,dj,ze,zzsj from myhive.ods.prdata where pch='202407090102000' limit 100;

2.flink sql read table，Read out the data normally
select dj,ze,zzsj from hoodie_catalog.flink_hudi.EHR_ZYYZMX_PRE_COW;

3.use hive sql read data
use flink_hudi;
select dj,ze,zzsj from test_cow;

**Expected behavior**
Failed with exception java.io.IOException:org.apache.hadoop.hive.ql.metadata.HiveException: java.lang.ClassCastException: org.apache.hadoop.io.Text cannot be cast to org.apache.hadoop.io.DoubleWritable

**Environment Description**

* Hudi version : 0.14

* Spark version : 3.3

* Hive version : 2.3.9

* Hadoop version : 3.0

* Storage (HDFS/S3/GCS..) : HDFS

* Running on Docker? (yes/no) : no

**Additional context**

调试hive日志输出，看到读出数据格式为[2014-06-06,0.0,0.0]，但我在hive查询的sql语句（select dj,ze,zzsj from test_cow;）第一列是dj，是一个double类型字段，但在这里读出来的是分区的字段的值

hive调试输出[2014-06-06,0.0,0.0]这个格式数据的类名为：org.apache.hadoop.hive.ql.exec.ListSinkOperator，调用方法为
```
public void process(Object row, int tag) throws HiveException {
try {
LOG.info("row数据:{},tag:{},class:{}", row, tag, row.getClass().getName());
LOG.info("row对应数据类型：{}", inputObjInspectors);

ClassLoader classLoader = fetcher.getClass().getClassLoader();
if (classLoader != null) {
// 尝试获取资源URLs
try {
Enumeration urls = classLoader.getResources(fetcher.getClass().getName().replace('.', '/') + ".class");
while (urls.hasMoreElements()) {
URL url = urls.nextElement();
LOG.info("Class " + fetcher.getClass().getName() + " is loaded from: " + url);
}
} catch (Exception e) {
e.printStackTrace();
}
} else {
LOG.info("Class " + fetcher.getClass().getName() + " is loaded by the bootstrap class loader.");
}

res.add(fetcher.convert(row, inputObjInspectors[0]));
numRows++;
runTimeNumRows++;
} catch (Exception e) {
throw new HiveException(e);
}
}
```

**Stacktrace**

```Add the stacktrace of the error.```
2024-07-08T14:42:03,814 ERROR [62dfdf81-d91C-40f8-91a2-d84f761a3671 main] liDriver: Failed with exception java.io.IOException:org.apache.haoophive.ql.metadata.HiveException：java.lang.ClassCastException：org.apache.hadoop.io.Text cannot be cast to org.apache.hadoop.io.DoubleWriteable
java.io.IOException:org.apache.haoophive.ql.metadata.HiveException：java.lang.ClassCastException：org.apache.hadoop.io.Text cannot be cast to org.apache.hadoop.io.DoubleWriteable
at org.apache.hadoop.hive.ql.exec. FetchTask. fetch(FetchTask.java:165)
at org.apache.hadoop.hive.ql.Driver.getResults(Driver.java:2208)
at orgapachehadop-hivecL1.CLriver.processocaCr0Evpriver:ava:257)
at org.apache.hadoop.hive.cli.CliDriver.processLine(CliDriver.java:407)
at org.apache.hadoop.hive.cli.CliDriver.executeDriver(CliDriver.java:825)
at org.apache.hadoop.hive.cli.CliDriver.run(CliDriver.java:763)
at org.apache.hadoop.hive.cli.CliDriver.main(CliDriver. java:690)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)

Start with the reported Hive entry points FetchTask.java and ListSinkOperator.process, then trace how the partitioned Hudi table's columns are ordered during Hive reads. Reproduce the query against the shown Hudi table and verify that selecting dj, ze, and zzsj returns those fields with their declared types instead of the fqsj partition value.

Written by the indexing model from the issue text.