Implement operator `Cast` for backend `Gorgonia`

[clarkmcc]

Why is this operator needed?

Disclaimer: I'm new to this machine-learning business. I've developed a TensorFlow model that I've converted to an onnx model. The model looks like this:

_________________________________________________________________
 Layer (type)                Output Shape              Param #   
=================================================================
 embedding_2 (Embedding)     (None, 500, 64)           5312      
                                                                 
 bidirectional_2 (Bidirectio  (None, 500, 256)         197632    
 nal)                                                            
                                                                 
 time_distributed_2 (TimeDis  (None, 500, 3)           771       
 tributed)                                                       
                                                                 
=================================================================
Total params: 203,715
Trainable params: 203,715
Non-trainable params: 0
_________________________________________________________________

Implementation

https://github.com/onnx/onnx/blob/main/docs/Operators.md#cast

Link to existing material on the backend

Expected problems?

N/A

Tests

?