Allow in-place operations

In !20 (merged) the case of Reshape forward kernel is disccussed by @olivierbichler

I think Reshape Operator should not copy any memory! We could in forward() use setRawPtr() on output's Tensor to input's Tensor rawPtr.

At several occasions for now, memory is wasted in Aidge due to systematic differentiation between input and output Tensor. It would be very useful for computation optimization to allow effective use of in-place computation.

Here is a web page that more deeply talks about it.

added Missing label

added to epic &6 (closed)

mentioned in merge request !20 (merged)

made the issue visible to everyone

aidge_core!367 (merged) fix this issue

closed