From: @tronzhang
Reviewed-by: @gaoxiong1,@anyrenwei
Signed-off-by: @anyrenwei

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

From: @alouhahahahaha
Reviewed-by: @jjfeing
Signed-off-by: @jjfeing

From: @liu_xiao_93
Reviewed-by: @liangchenghui
Signed-off-by: @liangchenghui

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

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

From: @dayschan
Reviewed-by: @gaoxiong1,@ckey_dou
Signed-off-by: @ckey_dou

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

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

1. Removed the deprecated pass "EliminateGetitemForControlDepend"
2. Spread the MakeTuple in UpdateState's input at PreProcess, so that all inputs are directly connected
   to UpdateState, I dont need to consider the condition "Getitem-MakeTuple-UpdateState'.
   after this pass, the UpdateState(U, make_tuple(op1, op2, ...)) was changed to UpdateState(U, op1, op2, ...)
3. Shrink the UpdateState's inputs at PostProcess. The reverse operation of the above pass.
   recovered the UpdateState's format for the process after GraphKernel.
4. Add a pass ExtendOutputForUpdateState, it's the main job of this commit.
   Consider this situation:
   A Cast op has multiple users in a composite kernel, while it's also in the output list and connects to
     an external UpdateState. In the pass "ShapeOpsSplitter", it will be duplicated. after that, only one replica will be connected
     to the external UpdateState, others will be connected to its original users respectively.
   After the pass "GraphKernelSplitter", only one part will be connected to this UpdateState, the execution order of other nodes cannot be ensured.
   This pass extended the node that connects to UpdateState, if a node has an external UpdateState user, all outputs that depend on this node
     will be connected to this UpdateState. It may add many redundant edges, the next pass will handle it.
5. Add a pass MergeOutputForUpdateState after GraphKernelSplitter.
   if an UpdateState has multiple inputs from the same node, only one edge will be kept.

From: @chuckchen521
Reviewed-by: 
Signed-off-by:

From: @liu_xiao_93
Reviewed-by: @liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui

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

From: @yuchaojie
Reviewed-by: @zhoufeng54,@jjfeing
Signed-off-by: @jjfeing

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

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

From: @liu_xiao_93
Reviewed-by: 
Signed-off-by:

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

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

From: @wilfchen
Reviewed-by: @cristoval,@limingqi107
Signed-off-by: @limingqi107

This bug make GetInputTensorNum() returns wrong number after communication_op_fusion.

From: @yuchaojie
Reviewed-by: @kisnwang
Signed-off-by:

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

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

From: @jjfeing
Reviewed-by: @stsuteng,@kisnwang
Signed-off-by: @stsuteng

From: @dayschan
Reviewed-by: @lingyunli63
Signed-off-by: