[SUPPORT]Same primary key with different _hoodie_record_key

[hbgstc123]

I have a table with a column video_id as primary key, and i find record with same primary key, have different record_key as shown in the picture below.

Steps to reproduce the behavior:
1.create table with spark sql, with tblproperties
tblproperties (
type = 'mor',
primaryKey = 'video_id'
)
2.insert historical data with spark sql
3.ingest real time incremental data with flink

config in flink ddl:

The _hoodie_record_key that written with spark contains a prefix "video_id:" while data written with flink doesn't

• Hudi version : 0.11.0

• Spark version : 3.1

• flink version : 1.13

• Storage (HDFS/S3/GCS..) : hdfs

• Running on Docker? (yes/no) : no

[danny0405]

This is a known problem, because Spark uses the ComplexAvroKeyGenerator by default even if the primary key only has one field, while flink would use SimpleAvroKeyGenerator instead when primary key fields is simple, a temporal solution is to set up the key generator for spark as SimpleAvroKeyGenerator instead manually, i would fire a fix for spark soon ~

[danny0405]

Have fired a fix here: #5815