Moving the CL entry for #9606 to the `v1.4.4` section. It's implied that anything in 1.n is also in 1.n+1
