[vLLM] add speculative configs

[lanking520]

Description

Initial PR to support speculative decoding

[lanking520]

Test on g5.12xl machine and everything is working fine

engine=Python
option.model_id=TheBloke/Llama-2-13B-fp16
option.speculative_draft_model=TinyLlama/TinyLlama-1.1B-Chat-v0.6
option.rolling_batch=vllm
option.tensor_parallel_degree=max
option.gpu_memory_utilization=0.8

curl -X POST http://127.0.0.1:8080/invocations -N   -H "Content-type: application/json"   -d '{"inputs":"tell me a story of the little red riding hood","parameters":{"max_new_tokens":256, "do_sample":true}}'
{"generated_text": " and mom im.\nshort little red riding hood, also available in sister's version, big bad wolf story: walt disney presents\nquick i need to quickly translate this story into english to learn.\ni need help, 'please can anybody help me in an essay about bubaris, and i dont know the origin of this story, or when/where it happened.\nplease can anybody help me in an essay about bubaris, and i dont know the origin of this story, or when/where it happened...can anybody help me?.\ncopied this off a site about sufism.. what do you think the story is about ?\nWho knows this Little Red Riding Hood story?\nI can't quite understand this sentence of a story, please help me.?\nExplain the story in these lyrics?\nHi, Could someone please help me translate this story about Elijah :-)\nI'm taking my story for school and i need a lot of examples please?\nHelp I need to make this story exciting?\nHelp, I need someone israel to speak to on a conference calling thing for a project on the holocaust. i have to"}

engine=Python
option.model_id=TheBloke/Llama-2-13B-fp16
option.speculative_draft_model=s3://djl-llm/tinyllama-1.1b-chat/
option.rolling_batch=vllm
option.tensor_parallel_degree=max
option.gpu_memory_utilization=0.8

The following settings also test passed