I'm not sure why, and I don't have sufficient python knowledge to figure
out. If anyone wants to give it a try please go ahead.

The just point to the existing implementation.

I'll try to add a test for it soon

In DebugMode, preemptive modif in case some op check if they have been applied in the Compute_map when their thunk is executed.

This optimization has been moved in sandbox/cuda

I've added another version of the optimization for gpu scans.

This import used to be on top of the file, but that creates a cycling
import.

preparing for the cython code, I used to grab the wrong inputs ( it should
be the node inputs not self.inputs)

reconstruct_graph always replaces CudaNdarrays with TensorTypes, so in case
of a gpu scan op, we can not use that logic to generate a hash ( I compute
it locally when I am generating the gpu scan op).

The way it works, __init__ gets a lambda function that construct a Tensor
Type. By default constructs Tensor Types, but move to gpu will replace it
with a CudaNdarray constructor.

Notes on this change: 1) I needed to replicate the inplace optimization, to
make a version that can handle ops moved on the gpu. I've also added a
replica of the reconstruct_graph function from scan_utils that again does
not replace CudaNdarray with TensorTypes ..

Two changes. First one is to make profiling work with scan (scan was looking
for a subclass of the ProfileStat object). Second, I made a more fixed
length printing of timings which I believe personally is much better.

While we agreed that there might be a more principial way of solving this,
this solution was fast to add and it is pretty efficient for now.

Is the shape of the data that should match.

In order for scan to run correctly inplace, it needs that none of the
initial states are the same memory buffer.