Add RL tutorial and run script for GPT-OSS 20B on Multi-Host TPUs

[susanbao]

Description

This PR adds a tutorial and a launch script for running Reinforcement Learning (RL) training with the GPT-OSS 20B model on Multi-Host TPUs.

Specifically, it:

• Creates the tutorial page docs/tutorials/posttraining/rl_gptoss_20b.md covering setup, checkpoint conversion (both to/from MaxText formats), and launching training on an GKE cluster with v5p-64 nodes.
• References the launch script scripts/run_gptoss_20b_rl.sh.
• Registers the new tutorial in docs/tutorials/post_training_index.md.

Tests

The tutorial is documentation and refers to existing code path definitions. The scripts/run_gptoss_20b_rl.sh file specifies standard run configurations for GPT-OSS 20B RL training.

Tested the script: LOG

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.
• [X ] 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

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!