Bug in gradient of fft.rfft

[ricardoV94]

Description

Reported in https://discourse.pymc.io/t/cannot-drop-a-non-broadcastable-dimension-and-rfft/14920

import pytensor.tensor as pt
from pytensor.tensor import fft

p = pt.vector("p", shape=(4,))
out = fft.rfft(p[None, :])
pt.grad(out.sum(), p)   # Cannot drop a non-broadcastable dimension: (False, False), [1]