Calculate the arithmetic mean of a single-precision floating-point strided array using ordinary recursive summation with extended accumulation and returning an extended precision result.