support alternative parallelism

[152334H]

--num-gpus is implemented by sharding each expert layer across GPUs, i.e. expert parallelism

this is probably not advisable for local experimentation, especially on batch size 1 -- where EP only adds communication overhead to no speed benefit vs naive model/pipeline parallel.

[tonysy]

Good suggestions, I am working on other parallelism method. Also, contribution is welcomed.