[torchlib] Fix wrong bias shape of ConvTranspose

[yuanyao-nv]

Previously, the bias shape of ConvTranpose was wrong since, unlike Conv, it should using the 1st dimension of weight shape and not the 0th. See described in #1299

In other words,

    if bias is None:
        weight_dim_0 = op.Shape(weight, start=0, end=1)
        bias_shape = op.Expand(weight_dim_0, op.Constant(value_ints=[1]))
        zero = op.CastLike(0.0, input)
        bias = op.Expand(zero, bias_shape)

should be changed to something like:

weight_dim_0 = op.Shape(weight, start=1, end=2) if transposed else op.Shape(weight, start=0, end=1)

However, I think it's more efficient to just eliminate bias altogether if it's not provided instead of filling it with zeros, since the ONNX spec allows bias to be absent.

[justinchuby]

Amazing, thanks!

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 75.09%. Comparing base (a7c797d) to head (afc0b5f).
Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1901   +/-   ##
=======================================
  Coverage   75.08%   75.09%           
=======================================
  Files         252      252           
  Lines       27417    27412    -5     
  Branches     3190     3189    -1     
=======================================
- Hits        20587    20584    -3     
+ Misses       5880     5878    -2     
  Partials      950      950           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.