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

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

From: @zhengjun10
Reviewed-by: 
Signed-off-by:

1. Search call info before insert nodes for control flow calls;
2. Use label_switch for multi return points if required.

From: @dessyang
Reviewed-by: @robingrosman,@linqingke
Signed-off-by: @linqingke

From: @yangruoqi713
Reviewed-by: @zhang_xue_tong,@zhanghaibo5
Signed-off-by: @zhang_xue_tong

From: @TFbunny
Reviewed-by: @tom__chen,@tom__chen,@robingrosman
Signed-off-by: @robingrosman

From: @peilin-wang
Reviewed-by: @liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui

2. Dump the end graph in gpu session and cu session

From: @zhaozhenlong
Reviewed-by: @zhang_xue_tong,@zhanghaibo5
Signed-off-by: @zhang_xue_tong

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

From: @zhangbuxue
Reviewed-by: @ginfung
Signed-off-by:

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

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

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

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

From: @xutianchun
Reviewed-by: @hangangqiang,@zhang_xue_tong
Signed-off-by: @hangangqiang

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

From: @Somnus2020
Reviewed-by: 
Signed-off-by:

From: @zh_qh
Reviewed-by: @ginfung,@ginfung,@hwhewei,@kingxian
Signed-off-by: @ginfung,@zhunaipan

refined comments

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

From: @alexyuyue
Reviewed-by: @robingrosman,@nsyca
Signed-off-by: @robingrosman

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

From: @anancds
Reviewed-by: @cristoval
Signed-off-by:

Signed-off-by: alex-yuyue <yue.yu1@huawei.com>

From: @anancds
Reviewed-by: @wilfchen
Signed-off-by: