From: @shenwei41
Reviewed-by: 
Signed-off-by:

From: @laiyongqiang
Reviewed-by: 
Signed-off-by:

From: @liubuyu
Reviewed-by: @kisnwang,@zhoufeng54
Signed-off-by: @zhoufeng54

From: @linqingke
Reviewed-by: @jjfeing,@xu-yfei
Signed-off-by: @xu-yfei

From: @wenfangpei
Reviewed-by: 
Signed-off-by:

From: @laiyongqiang
Reviewed-by: @kisnwang,@zhoufeng54
Signed-off-by: @zhoufeng54

From: @zhoufeng54
Reviewed-by: @xu-yfei,@kisnwang
Signed-off-by: @xu-yfei

model converter

Signed-off-by: zhoufeng <zhoufeng54@huawei.com>

From: @liubuyu
Reviewed-by: @kisnwang,@zhoufeng54
Signed-off-by: @zhoufeng54

2. fix bug of bounding_box_encode kernel_name

Support Dynamic Shape Aicpu Run Package

adapt tensorengin modify, fix ub fusion

- It provides a unified style to express graph and kernel for user.
- It provides a unified IR to represent graph and kernel for developer.
- It breaks the boundary between graph and kernel.
- It provides more opportunities to do compile optimization.

iou

NMSWithMask

larsupdate testcase

sgd testcase

Previously function op_select_format and check_supported raise an exception directly on the tbe_process python side, but we don't deal with the exception, and raise an exeception on c++ side to frontend ME, that will cause some conflict when recycle resource on ME and tbe_process python interpreter.
This changes adding try...catch in function op_select_format and check_supported on the python side, and return the Exception string to c++ side, so that we can raise an exception to frontend ME and ME will deal with resouce clearning and exit.