Add a new DAG for SFT post-training from Jupyter notebooks

[ooops678]

Description

Airflow DAG for automating Llama3.1-8B SFT training from Jupyter notebook.
This DAG automates the sft_llama3_demo.ipynb notebook, executing SFT
training on single-host TPU VMs.

Tests

• Airflow Test Results: https://f5eac1c8d1684611864003492b988589-dot-us-east1.composer.googleusercontent.com/dags/maxtext_sft_notebook/grid?task_id=maxtext_sft_notebook_sft-v6e-8.run_model&dag_run_id=manual__2026-01-27T07%3A37%3A11.174725%2B00%3A00

Airflow/Composer

• GCP Composer name: camilo-orbax-dev (under GCP project: cloud-ml-auto-solutions)
• GCP Composer version: 2.13.1

Bucket with HNS (Hierarchical Name Space)

• Bucket Name: gs://mtc-automation-bucket

Checklist

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

• I have performed a self-review of my code.
• I have necessary comments in my code, particularly in hard-to-understand areas.
• I have run one-shot tests and provided workload links above if applicable.
• I have made or will make corresponding changes to the doc if needed.