[HUDI-3922] parse record key + partition path config consistently between keygens and HiveSync

[qjqqyy]

What is the purpose of the pull request

hive-sync and the key generators (such as ComplexKeyGenerator) handles the degenerate case (empty string) differently.
Previously this was merely a configuration annoyance, but now it causes issues (see #5101).

We parse configs consistently between the various key generators and HiveSyncConfig

Brief change log

• Filter empty strings in TypedProperties#getStringList (this causes HoodieSyncConfig to treat hoodie.datasource.hive_sync.partition_fields= the same as not specifying it)
• Parse PARTITIONPATH_FIELD_NAME similarly in the various key generators
  • in particular, it now accepts both not specifying the prop key and setting it as the empty string to denote a non-partitioned table
• While here, also parse RECORDKEY_FIELD_NAME similarly, and allow leaving not specifying RECORDKEY_FIELD_NAME to mean the default.

Verify this pull request

This PR modifies the handling of degenerate configurations. The affected test cases are updated to match the new behaviour.

Committer checklist

• Has a corresponding JIRA in PR title & commit

• Commit message is descriptive of the change

• CI is green

• Necessary doc changes done or have another open PR

• For large changes, please consider breaking it into sub-tasks under an umbrella JIRA.

[hudi-bot]

CI report:

• 32733bd Azure: SUCCESS

Bot commands

@hudi-bot supports the following commands:

• @hudi-bot run azure re-run the last Azure build

[codope]

@qjqqyy Thanks for putting up a fix. I have some concern as mentioned below. I would suggest to not make any change in the keygen logic, and isolate the fix to hive-sync module only.