fix: Repair Citrinet-1024 compilation issues [Duplicate of PR #1488 for Release 1.3]

[gs-olive]

Description

Duplicate of PR #1488 targeted at Release 1.3
Fix compilation issues with Citrinet-1024 arising from type-casting issue in aten::sum and layer-naming issue in aten::div.

• Enable automatic type-casting in aten::sum for bool tensor inputs to agree with Torch casting behavior
• Fix bug in aten::div where all internal div layers have the same name
• Add test cases for aten::sum type-casting

Without type-casting, the error arises from the following:

DEBUG: [Torch-TensorRT - Debug Build] - ITensor shape: [1, 1, 256]
DEBUG: [Torch-TensorRT - Debug Build] - ITensor type: Bool
DEBUG: [Torch-TensorRT - Debug Build] - InDims 1 1 256
DEBUG: [Torch-TensorRT - Debug Build] - Dim to reduce(original):[-1]
DEBUG: [Torch-TensorRT - Debug Build] - Dim to reduce(converted):[2]
DEBUG: [Torch-TensorRT - Debug Build] - Axis Mask: 00000000000000000000000000000100
DEBUG: [Torch-TensorRT - Debug Build] - Keep dims: 1
WARNING: [Torch-TensorRT - Debug Build] - Sum converter disregards dtype
ERROR: [Torch-TensorRT TorchScript Conversion Context] - 3: %66 : Tensor = aten::sum(%mask2.2, %65, %26, %3779)
# ReduceLayer only supports Float, Half, Int8, and Int32 data types.

With type casting, one error is resolved, but another appears:

ERROR: [Torch-TensorRT TorchScript Conversion Context] - 4: [network.cpp::validate::2761] Error Code 4: Internal Error (Repeated layer name: tmp_div (layers must have distinct names))
ERROR: [Torch-TensorRT TorchScript Conversion Context] - 2: [builder.cpp::buildSerializedNetwork::742] Error Code 2: Internal Error (Assertion engine != nullptr failed. )

With both updates, the model compiles succesfully.

Fixes #1487

Type of change

• Bug fix (non-breaking change which fixes an issue)

Checklist:

• [ x ] My code follows the style guidelines of this project (You can use the linters)
• [ x ] I have performed a self-review of my own code
• [ x ] I have commented my code, particularly in hard-to-understand areas and hacks
• [ x ] I have made corresponding changes to the documentation
• [ x ] I have added tests to verify my fix or my feature
• [ x ] New and existing unit tests pass locally with my changes
• [ x ] I have added the relevant labels to my PR in so that relevant reviewers are notified

[narendasan]

Looks Good