Separate RL configurations into a dedicated config class

[SurbhiJainUSC]

Description

This PR separates RL configurations into its own dedicated configuration class.

Why is this change being made & the problem being solved:
Previously, RL-specific configurations were mixed with the pre-training configurations. As the number of RL-specific features and hyperparameters grows, this mixed approach makes the primary configuration structure bloated and harder to maintain.

Why this is a good solution:
By isolating RL configurations into a dedicated class, we improve the modularity and readability of our configuration management. It enforces a clear separation of concerns, ensuring that non-RL workloads aren't cluttered with RL-specific parameters, and makes RL hyperparameters easier to track and modify independently.

Specific implementation:

• Created a new dedicated config class specifically for RL parameters.
• Migrated existing RL configuration variables out of the base config and into this new class.
• Updated configuration references across the codebase to point to the newly established RL config class.

Tests

Command: https://paste.googleplex.com/5172847176712192

Checklist

Before submitting this PR, please make sure (put X in square brackets):

• I have performed a self-review of my code. For an optional AI review, add the gemini-review label.
• I have necessary comments in my code, particularly in hard-to-understand areas.
• I have run end-to-end tests tests and provided workload links above if applicable.
• I have made or will make corresponding changes to the doc if needed, including adding new documentation pages to the relevant Table of Contents (toctree directive) as explained in our documentation.

[codecov]

Codecov Report

❌ Patch coverage is 27.50000% with 58 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/maxtext/models/models.py	22.22%	0 Missing and 14 partials ⚠️
src/maxtext/utils/maxtext_utils.py	41.66%	12 Missing and 2 partials ⚠️
src/maxtext/layers/nnx_decoders.py	28.57%	4 Missing and 1 partial ⚠️
src/maxtext/common/metric_logger.py	0.00%	2 Missing and 1 partial ⚠️
src/maxtext/utils/model_creation_utils.py	40.00%	3 Missing ⚠️
...ation/vllm/torchax_converter/validate_converter.py	0.00%	2 Missing ⚠️
src/maxtext/layers/quantizations.py	50.00%	1 Missing and 1 partial ⚠️
src/maxtext/layers/decoders.py	0.00%	1 Missing ⚠️
src/maxtext/models/deepseek.py	0.00%	0 Missing and 1 partial ⚠️
src/maxtext/models/gemma.py	0.00%	1 Missing ⚠️
... and 12 more

📢 Thoughts on this report? Let us know!

[github-actions]

🤖 Hi @SurbhiJainUSC, I've received your request, and I'm working on it now! You can track my progress in the logs for more details.

[github-actions]

🤖 I'm sorry @SurbhiJainUSC, but I was unable to process your request. Please see the logs for more details.