feat: skip validation and reuse pre-converted arrow cache

[macandro96]

What does this PR do ?

• Load NeMo Gym JSONL through Hugging Face’s reusable Arrow cache while preserving raw rows and supporting pre-converted Arrow/Parquet datasets; observed dataset setup improved from ~152 seconds to 2.2 seconds.

• Skip validation dataset loading when validation is disabled, avoiding unnecessary duplicate data initialization while preserving validation and trajectory-collection behavior.

Issues

List issues that this PR closes (syntax):

Usage

• You can potentially add a usage example below

# Add a code snippet demonstrating how to use this

Before your PR is "Ready for review"

Pre checks:

• Make sure you read and followed Contributor guidelines
• Did you write any new necessary tests?
• Did you run the unit tests and functional tests locally? Visit our Testing Guide for how to run tests
• Did you add or update any necessary documentation? Visit our Document Development Guide for how to write, build and test the docs.

Additional Information

• ...

[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.

[macandro96]

/ok to test 6aff5f9