Add support for UINT32 for ttnn.copy/typecast

[ayerofieiev-tt]

Running bert in PyTorch 2.0

ttnn.copy only supports float/bfloat inputs but got DataType::UINT32

full log

return self.function(*function_args, **function_kwargs)
RuntimeError: TT_FATAL @ /work/ttnn/cpp/ttnn/operations/data_movement/copy/device/copy_device_operation.cpp:19: input_tensor_a.get_dtype() == DataType::BFLOAT16 or input_tensor_a.get_dtype() == DataType::BFLOAT8_B or input_tensor_a.get_dtype() == DataType::FLOAT32 or input_tensor_a.get_dtype() == DataType::BFLOAT4_B
info:
ttnn.copy only supports float/bfloat inputs but got DataType::UINT32
backtrace:
...
 --- ttnn::operations::data_movement::CopyDeviceOperation::validate_with_output_tensors(std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>> const&, std::__1::vector<std::__1::optional<tt::tt_metal::Tensor>, std::__1::allocator<std::__1::optional<tt::tt_metal::Tensor>>> const&) const
 --- /home/artem/demo/python_env/lib/python3.10/site-packages/ttnn/_ttnn.cpython-310-x86_64-linux-gnu.so(_ZN4ttnn16device_operation6detail23launch_on_worker_threadIN2tt8tt_metal9operation23OldInfraDeviceOperationINSt3__16vectorINS4_6TensorENS7_9allocatorIS9_EEEEEENS3_3stl10StrongTypeIhNS_10QueueIdTagEEElNS5_15DeviceOperationISC_EENSD_13tensor_args_tESC_PNS4_2v07IDeviceEEEvT0_T1_RKT2_RKT3_RT4_RT5_+0x23e) [0x7df0f4356a9e]
 --- /home/artem/demo/python_env/lib/python3.10/site-packages/ttnn/_ttnn.cpython-310-x86_64-linux-gnu.so(_ZN4ttnn16device_operation6detail23launch_on_single_deviceIN2tt8tt_metal9operation23OldInfraDeviceOperationINSt3__16vectorINS4_6TensorENS7_9allocatorIS9_EEEEEEEENT_21tensor_return_value_tENS3_3stl10StrongTypeIhNS_10QueueIdTagEEERKNSE_22operation_attributes_tERKNSE_13tensor_args_tE+0x83) [0x7df0f43567b3]
 --- /home/artem/demo/python_env/lib/python3.10/site-packages/ttnn/_ttnn.cpython-310-x86_64-linux-gnu.so(+0x2f566aa) [0x7df0f43566aa]
 --- /home/artem/demo/python_env/lib/python3.10/site-packages/ttnn/_ttnn.cpython-310-x86_64-linux-gnu.so(_ZN4ttnn16device_operation6detail6invokeIN2tt8tt_metal9operation23OldInfraDeviceOperationINSt3__16vectorINS4_6TensorENS7_9allocatorIS9_EEEEEEEENT_21tensor_return_value_tENS3_3stl10StrongTypeIhNS_10QueueIdTagEEERKNSE_22operation_attributes_tERKNSE_13tensor_args_tE+0x197) [0x7df0f4356067]
...
 --- std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>> tt::tt_metal::operation::run<std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>>>(tt::tt_metal::operation::DeviceOperation<std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>>>&&, std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>> const&, std::__1::vector<std::__1::optional<tt::tt_metal::Tensor const>, std::__1::allocator<std::__1::optional<tt::tt_metal::Tensor const>>> const&, std::__1::vector<std::__1::optional<tt::tt_metal::Tensor>, std::__1::allocator<std::__1::optional<tt::tt_metal::Tensor>>> const&, tt::stl::StrongType<unsigned char, ttnn::QueueIdTag>)
 --- /home/artem/demo/python_env/lib/python3.10/site-packages/ttnn/_ttnn.cpython-310-x86_64-linux-gnu.so(+0xeb5287) [0x7df0f22b5287]
 --- ttnn::operations::experimental::copy::TypecastOperation::invoke(tt::stl::StrongType<unsigned char, ttnn::QueueIdTag>, tt::tt_metal::Tensor const&, tt::tt_metal::DataType const&, std::__1::optional<tt::tt_metal::MemoryConfig> const&, std::__1::optional<tt::tt_metal::Tensor> const&)
...
 --- void tt::tt_metal::operation::launch_op_func<std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>>>(std::__1::function<std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>> (std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>> const&, std::__1::vector<std::__1::optional<tt::tt_metal::Tensor const>, std::__1::allocator<std::__1::optional<tt::tt_metal::Tensor const>>> const&, std::__1::vector<std::__1::optional<tt::tt_metal::Tensor>, std::__1::allocator<std::__1::optional<tt::tt_metal::Tensor>>> const&)> const&, std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>>, std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>>&, std::__1::vector<std::__1::optional<tt::tt_metal::Tensor const>, std::__1::allocator<std::__1::optional<tt::tt_metal::Tensor const>>>, std::__1::vector<std::__1::optional<tt::tt_metal::Tensor>, std::__1::allocator<std::__1::optional<tt::tt_metal::Tensor>>>, bool)
...
 --- void tt::tt_metal::operation::launch_op_func<std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>>>(std::__1::function<std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>> (std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>> const&, std::__1::vector<std::__1::optional<tt::tt_metal::Tensor const>, std::__1::allocator<std::__1::optional<tt::tt_metal::Tensor const>>> const&, std::__1::vector<std::__1::optional<tt::tt_metal::Tensor>, std::__1::allocator<std::__1::optional<tt::tt_metal::Tensor>>> const&)> const&, std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>>, std::__1::vector<tt::tt_metal::Tensor, std::__1::allocator<tt::tt_metal::Tensor>>&, std::__1::vector<std::__1::optional<tt::tt_metal::Tensor const>, std::__1::allocator<std::__1::optional<tt::tt_metal::Tensor const>>>, std::__1::vector<std::__1::optional<tt::tt_metal::Tensor>, std::__1::allocator<std::__1::optional<tt::tt_metal::Tensor>>>, bool)
...

[ayerofieiev-tt]

@rdjogoTT , do you know if there any reason why typecast does not support uint32?