Function reduce 

pub fn reduce<T: Copy + MaybeSendSync, Op: ElementOp<T>, M, R, U>(
    src: &StridedView<'_, T, Op>,
    map_fn: M,
    reduce_fn: R,
    init: U,
) -> Result<U>
where
    M: Fn(T) -> U + MaybeSync,
    R: Fn(U, U) -> U + MaybeSync,
    U: Clone + MaybeSendSync,

Full reduction with map function: reduce(init, op, map.(src)).

Evaluation order

reduce_fn must be associative and commutative, and init must be an identity of reduce_fn whenever the reduction runs on several threads. A contiguous source is folded with sixteen independent lanes (see the crate source for the exact lane assignment), non-contiguous sources are traversed in a cache-friendly loop order, and parallel execution folds chunks independently and combines them in chunk order. Floating point results can therefore differ from a strict left to right fold in the last bits; for a fixed shape, layout and thread count the result is repeatable.

Because map_fn and reduce_fn are opaque closures this entry cannot select the dtype-specific SIMD kernels of crate::ErasedReducePlan; callers that reduce with a known operation (sum, product, max, min) on a supported dtype can use that plan for the fastest path.