Update TimeSeries stage to also work with Production DFP

[efajardo-nv]

Description

• Update TimeSeries stage to also work with Production DFP pipeline instead of Starter DFP. Now accepts MultiMessage instead of MultiResponseMessage. Remove use of hard-coded event_dt and replace with config.ae.timestamp_column_name.
• Tested updated TimeSeries stage with Duo and Azure production DFP pipelines.
• As part of a Starter CloudTrail DFP pipeline, TimeSeries stage runs into issues when processing multiple batches, i.e. when repeat option in CloudTrailSourceStage is increased (increase input data size) or pipeline batch size is decreased. Pipeline either errors out or not all data is processed. Remove TimeSeries stage from AutoEncoder E2E benchmark for now and will create separate issue for this.

Fixes #1107

Checklist

[x] I am familiar with the Contributing Guidelines.
[x] New or existing tests cover these changes.
[x] The documentation is up to date with these changes.

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[efajardo-nv]

/ok to test

[efajardo-nv]

/merge