Multiplication of vector with scalar is a common operation. I also find the syntax of mul function to be nicer than creating the vector of the required size (mul(v, s) vs. v * f32x4s(s)). Also, it seems natural to try to use this function for it.

If you agree, I could prepare PR.