Disable runtime broadcasting in indexing operations

Description

We are inconsistent in what Ops we allow runtime broadcasting and which we don't.

import pytensor.tensor as pt

x = pt.vector("x", shape=(None,))  # Not known to have length 1 at runtime
out = pt.alloc(x, 3, 5)
try:
    out.eval({x: [1]})
except Exception as e:
    print(str(e).splitlines()[0])
# Runtime broadcasting not allowed. The output of Alloc requires broadcasting a dimension of the input value, which was not marked as broadcastable. If broadcasting was intended, use `specify_broadcastable` on the relevant input.

out = pt.zeros((10, 10))[[5, 6, 7], [0, 1, 2]].inc(x)
try:
    out.eval({x: [1]})
except Exception as e:
    print(str(e).splitlines()[0])
else:
    print("Did not raise")
# Did not raise

Note that whenever we allow runtime broadcasting will have a wrong gradient wrt to that broadcasted input, since we never implemented a mechanism to reduce runtime broadcasted dimensions.

print(pt.grad(out.sum(), x).eval({x: [1]}).shape)  # (3,)

Start by reproducing the two indexing examples in the issue and tracing the indexing operation implementations and their gradient handling. Compare their runtime broadcasting checks with pt.alloc, then verify that indexing operations reject unmarked runtime broadcasting and no longer produce an incorrect gradient for that case.