✨[Feature] + 🐛 [Bug] Allow ITensor biases in aten.convolution converters

[gs-olive]

Context

Currently, aten.convolution converters do not support ITensor biases, which can cause test failures in CI (example), as the new Dynamo compile path primarily uses ITensor objects for general tensors throughout computation.

TensorRT/py/torch_tensorrt/fx/converters/acc_ops_converters.py

Lines 202 to 208 in d3a47c4

	# and bias being ITensor is not supported in TensorRT api
	# right now
	if kwargs["bias"] is not None and not isinstance(kwargs["bias"], torch.Tensor):
	raise RuntimeError(
	f"linear {name} has bias of type {type(kwargs['bias'])}, Expect Optional[Tensor]"
	)
	bias = to_numpy(kwargs["bias"]) # type: ignore[arg-type]

Proposed Solution

Allow ITensor biases for aten.convolution ops in the same way that the kernel weights can be ITensor objects. See IConvolutionLayer for further information on TensorRT convolution layers.

[frank-wei]

QQ, what is the ITensor here?

[gs-olive]

The ITensor is a TRTTensor representing the bias term in the convolution. It arises in a few Dynamo export/compile cases where the bias tensor is registered as an input to a subgraph during functionalization.