Forward input dims in Reshape ShapeHelper

[tungld]

In this reshape op in a quantized bert model, the first two dimensions of the output are the same as the ones in the input, so we would like to forward IndexExprs of these dimensions from the input to the output so that DimAnalysis works. But it failed because of using IndexExpr.selectOrSelf.

%184 = onnx.Constant dense<[0, 0, 12, 64]> : tensor<4xi64>
%494 = "onnx.Reshape"(%491, %184) {allowzero = 0 : si64, onnx_node_name = "/bert/encoder/layer.0/attention/self/Reshape"} : (tensor<?x?x768xf32>, tensor<4xi64>) -> tensor<?x?x12x64xf32>

I think the better solution would be updating selectOrSelf, but I don't really understand it and just make a change in the Reshape's ShapeHelper as a workaround.

[tungld]

@AlexandreEichenberger if you know how to modify selectOrSelf to fix this issue, please let me know or please feel free to create a PR for that. Thanks!

[tungld]

I changed this to make the compare expression literal.

[tungld]

It looks like I found the way to change selectOrSelf but it only works if the condition is literal at compile time.

[AlexandreEichenberger]

@tungld is there still a case to look into?

[tungld]

@tungld is there still a case to look into?

I changed selectOrSelf to handle the case where the comparison is literal. It works for the quantized bert model. I hope it's enough for most cases.

[tungld]

@AlexandreEichenberger It's ready for review.