Specialize PadUsing::[r]fold

[Philippe-Cholet]

Related to #755

cargo bench --bench specializations "pad_using/r?fold"

pad_using/fold          [1.2169 µs 1.2273 µs 1.2375 µs]
pad_using/fold          [679.64 ns 683.21 ns 687.65 ns]
                        [-45.562% -44.840% -44.185%]

pad_using/rfold         [1.3184 µs 1.3243 µs 1.3317 µs]
pad_using/rfold         [692.28 ns 694.11 ns 695.98 ns]
                        [-47.743% -47.421% -47.084%]