* changed '/ppml' into '/bigdl' and modified llama-7b
* Added the contents of finetuning in README
* Modified link of qlora_finetuning.py in Dockerfile
* add instruction for MacOS/Linux
* modify html label of gif images
* organize structure of README
* change title name
* add inference-xpu, serving-cpu and serving-xpu parts
* revise README
* revise README
* revise README