Use non-deprecated PyTorch methods to silence warnings (#541)

Getting warnings of the following form using ToT TE:

```
/usr/local/lib/python3.10/dist-packages/transformer_engine/pytorch/attention.py:852: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly.  To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
   data_ptr = grad_outputs[0].storage().data_ptr()
```
Signed-off-by: Deepak Narayanan <2724038+deepakn94@users.noreply.github.com>

transformer_engine/pytorch/attention.py

@@ -850,7 +850,7 @@ class _SplitAlongDim(torch.autograd.Function):
 
         noop_ok = True
         strides = grad_outputs[0].stride()
-        data_ptr = grad_outputs[0].storage().data_ptr()
+        data_ptr = grad_outputs[0].untyped_storage().data_ptr()
         shape = list(grad_outputs[0].shape)
         for i, tensor in enumerate(grad_outputs):
             shape_i = shape
@@ -858,7 +858,7 @@ class _SplitAlongDim(torch.autograd.Function):
             offset_size = sum(split_sizes[:i]) * np.prod(shape[split_dim+1:])
             if (tensor.stride() != strides or
                 list(tensor.shape) != shape_i or
-                tensor.storage().data_ptr() != data_ptr or
+                tensor.untyped_storage().data_ptr() != data_ptr or
                 tensor.storage_offset() != offset_size):
                 noop_ok = False
                 break