- Add floating point number data type, with two possible precisions
    - single precision : like Rust's f32
    - double precision : like Rust's f64