How to perform explicit broadcasting in dynet ?

I have a batched input Tensor of shape : T * d * b where b is the batched dimension. I need to divide this by another Tensor : d * b . Function **cdiv** needs either the same dimension or unmatched dimension on the right. How can I manipulate tensors to perform this functionality?

No source files, tests, or entry points are named. Start by locating DyNet's cdiv tensor operation and its shape and broadcasting rules; clarify whether the expected outcome is an existing tensor manipulation or a new explicit-broadcasting capability, then verify handling of the stated T*d*b divided by d*b shapes.

Written by the indexing model from the issue text.