Let's change shmem_alloc_folio() to take a order and use
folio_alloc_mpol() helper, then directly use it for normal or large folio
to cleanup code.

Link: https://lkml.kernel.org/r/20240515070709.78529-5-wangkefeng.wang@huawei.comSigned-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>