Specialize TakeWhileInclusive::fold

[Philippe-Cholet]

Related to #755

cargo bench --bench specializations "take_while_inclusive/fold"

take_while_inclusive/fold   [556.44 ns 559.37 ns 562.70 ns]
                            [536.51 ns 538.17 ns 539.76 ns]
                            [-6.2459% -4.9270% -3.8918%]

It's not much of a win but try_fold does not show real performance wins on slices and we benchmark on slices. It would perform better on other kind of iterators.