Add Llama2-70b sparsecore collective model to trillium configs

[Obliviour]

Description

Add a new llama2-70b config with sparsecore offloading that scales up to 32 pods so far. Adjust the num_slices arg below:

1. build maxtext dependencies
2. have xpk in ~/xpk/xpk.py
3. run below command

Tests

• Ran successful training on TPUs

python3 benchmarks/benchmark_runner.py --project=${PROJECT} --zone={zone} --device_type=v6e-256 --num_slices=1 --cluster_name=${CLUSTER_NAME} --base_output_directory=${OUTPUT_DIR}
--model_name="llama2_70b_4096_sc" --libtpu_version=20241106 --base_docker_image=maxtext_base_image

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 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.