Sort a Vector of Floats - total_cmp vs. partial_cmp

[Hasnabruzzn]

In Sort a Vector of Floats partial_cmp is used, which panics with NaN.

For cases where a vector of floats can contain NaNs there is total_cmp (since Rust 1.64?)

    let mut vec = vec![1.1, 1.15, 5.5, 1.123, 2.0, f32::NAN];

    vec.sort_by(|a, b| a.total_cmp(b));

    assert_eq!(vec[0..5], vec![1.1, 1.123, 1.15, 2.0, 5.5]);

I don't know whether partial_cmp would be preferable if there aren't any NaN.