Speed up a little the slicing-by-8 code path by replacing the
(load+shift+xor)*4 sequence with a single u32 load plus a xor.
Before:
```
iterative: 1018 MiB/s [000000006c3b110d]
small keys: 1075 MiB/s [0035bf3dcac00000]
```
After:
```
iterative: 1114 MiB/s [000000006c3b110d]
small keys: 1324 MiB/s [0035bf3dcac00000]
```