docs(rl): update tutorial with AgenticGRPOLearner for async RL training

[AntonyMei]

Description

Updates the rl.md tutorial documentation to include instructions and an example command for using AgenticGRPOLearner for asynchronous RL training.

Tests

Command tested on Qwen3 4B using a v6e-8 machine.

# [Test Agentic Learner (default)]
export MODEL="qwen3-4b"
export BASE_OUTPUT_DIRECTORY="gs://yixuanm-maxtext-logs/AgenticGRPOTesting"
export RUN_NAME="agentic-0610-default"
export CHIPS_PER_VM=8
export MAXTEXT_CKPT_PATH="gs://yixuanm-maxtext-logs/AgenticGRPOTesting/qwen3-4b-ckpt/0/items/"

# GRPO command
python3 -m maxtext.trainers.post_train.rl.train_rl \
  model_name=${MODEL?} \
  load_parameters_path=${MAXTEXT_CKPT_PATH?} \
  run_name=${RUN_NAME?} \
  base_output_directory=${BASE_OUTPUT_DIRECTORY?} \
  chips_per_vm=${CHIPS_PER_VM?} \
  enable_tunix_perf_metrics=True \
  rl.use_agentic_rollout=True 

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.