Optimise the decomposition for `adaptive_avg_pool2d` wrt. TorchInductor (#84483)
This fixes some part of the implementation that did not work with
TorchInductor (e.g. the indices in TorchInductor need to be `int64`s,
while in PyTorch we can have `int32`s).
It also brings up the performance of the kernel to similar numbers than
those of the lowering (benchmarks below).
Pull Request resolved: https://github.com/pytorch/pytorch/pull/84483
Approved by: https://github.com/jansel
