Wrap up the distributed benchmark and get them running on conda_mast

[xuzhao9]

We would like to introduce basic distributed benchmarking support on synthetic data.

The idea is to wrap up single-GPU model on DDP/FSDP, then get them running on conda_mast. The initial OSS distributed userbenchmark could be used as the starting point: https://github.com/pytorch/benchmark/tree/main/userbenchmark/distributed