Docs migration from XPK to CTK(3/5)

[KevinYenky]

Description

This PR is the third part of decomposing the XPK to Cluster Toolkit document migration (from PR #5175). It focuses on updating our post-training tutorial documentation for Reinforcement Learning to use Cluster Toolkit (CTK) instead of the deprecated XPK launcher.

Specifically, this PR updates the following RL guides:

• Multi-Host RL (GRPO & GSPO): Replaces xpk workload create-pathways with gcluster job submit for both GRPO and GSPO workloads, updating environment setup variables (like COMPUTE_TYPE and TOPOLOGY) and job monitoring commands.
• Model-Specific RL Tutorials (Gemma 4, GPT-OSS 20B, Qwen3 30B): Updates the step-by-step submission instructions to use CTK and gcluster commands instead of xpk scripts.
• RL README: Updates prerequisite instructions to recommend Cluster Toolkit for standard GKE workloads over XPK.

Why this is being made:
CTK is replacing XPK as the primary submission mechanism for Google Kubernetes Engine (GKE) workloads. Breaking down the massive PR #5175 into smaller, logical chunks (starting with these core guides) makes the migration significantly easier to review and test.

Tests

tested by scripts that run the gcluster command provided in the documents, results stored in gcs bucket.

Checklist

• 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.

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[gemini-code-assist]

Code Review

This pull request updates several post-training reinforcement learning (RL) tutorials to use Cluster Toolkit (gcluster) instead of XPK for cluster management and job submission. Feedback on these changes includes correcting the order of export commands in the retry instructions of rl_on_multi_host.md to prevent directory resolution issues, specifying the train_1M subset for the large nvidia/OpenMathInstruct-2 dataset in both the Gemma 4 and Qwen 3 tutorials to avoid memory timeouts, and aligning the subslice parameters in the experimental RL README with the described device allocations.