Function shift 

pub fn shift<T: RudaPrimitive>(
    input: &Array<T>,
    output: &mut Array<T>,
    scratch: &mut SharedMemory<T>,
    threads: usize,
    items_per_thread: usize,
    down: bool,
) -> T

Shift a blocked register tile one item. Boundary output is preserved. Returns the block’s discarded prefix (down) or suffix (up).