[LITE][NPU][XPU] Refine subgraph pass, and support NPU/XPU model generation at execution time (#2576)

* add static library in tiny_publish
* move flto and ffunction-sections cmake option into the tiny publish so result of java,cxx,python 

* Add supporting for Huawei offical DDK
* Fix the param of graph op in NPU graph computing kernel

* [framework][place] remove prefered_place, use place order in valid_place array instead test=develop

* remove kHost from valid_places test=develop

* [NPU] add NPU supporting for Java API
test=develop

* [NPU] refine build script for NPU compiling
test=develop

* [NPU] fix compiling script for NPU
test=develop

* [Publish] fix publish compile

* update build.sh to clean build dir