vdk-oracle: ORA-01002: fetch out of sequence error in _cache_tables when some rows fail to ingest

[DeltaMichael]

Overview

While doing load testing, this error was happening sporadically and failing for 1% of ingestion rows. It might be due to unnecessary commit calls.

https://github.com/vmware/versatile-data-kit/blob/main/projects/vdk-plugins/vdk-oracle/src/vdk/plugin/oracle/ingest_to_oracle.py#L164

We probably don't need this commit call, because of commit being called on connection close anyway.

https://github.com/vmware/versatile-data-kit/blob/main/projects/vdk-plugins/vdk-oracle/src/vdk/plugin/oracle/oracle_connection.py#L56

This has to be tested, however.

How to reproduce

1. Remove this try-except
2. Ingest 1 000 000 records using a functional test where you generate the table id in python code, e.g.

for i in range(1000000):
    p = payload.copy()
    p["id"] = i
    job_input.send_object_for_ingestion(
        payload=p, destination_table="test_table"
    )

3. Check result logs for ORA-01002

Acceptance criteria

1. Reproduce the bug and fix it
2. Remove the extra commit call and test that everything else works correctly