[SUPPORT] How to configure spark and flink to write mor tables using bucket indexes?

I want to use flink and spark to write to the mor table, and use bucket CONSISTENT_HASHING for the index, but I find that spark is very fast to write the full amount and flink is very slow(flink write 100record/s) to write increments.
spark sql：
```
CREATE TABLE test.tableA ()
USING hudi
TBLPROPERTIES (
'connector' = 'hudi',
'index.type'='BUCKET',
'hoodie.index.type'='BUCKET',
'hoodie.index.bucket.engine'='CONSISTENT_HASHING',
'hoodie.datasource.write.recordkey.field' = '',
'path' = '',
'preCombineField' = 'create_time',
'precombine.field' = 'create_time',
'primaryKey' = '',
'table.type' = 'MERGE_ON_READ',
'write.rate.limit'='10000',--flink配置
'write.tasks'='2',--flink配置
'write.utc-timezone'='false',
'type' = 'mor');
```
How to optimize？

Start with the supplied Spark SQL definition and Flink settings, focusing on the MOR table, CONSISTENT_HASHING bucket index, write rate limit, and task count. Reproduce a controlled incremental-write comparison and use the reported throughput to identify the limiting configuration or behavior; done means an evidence-based optimization or explanation for the Flink performance gap.

Written by the indexing model from the issue text.