needless_range_loop makes questionable suggestion

Pretty simple. Here is the code (I didn't write it):

    pub fn sum(&self) -> (Vec<f64>, Vec<f64>) {
        let mut sum_x = vec![0.0; self.x[0].len()];
        let mut sum_y = vec![0.0; self.y[0].len()];

        for i in 0..self.x.len() {
            for j in 0..self.x[i].len() {
                sum_x[j] += self.x[i][j];
            }
        }

        for i in 0..self.y.len() {
            for j in 0..self.y[i].len() {
                sum_y[j] += self.y[i][j];
            }
        }

        (sum_x, sum_y)
    }

self.x and self.y are Vec<Vec<f64>>.

Clippy says:

warning: the loop variable `j` is used to index `sum_x`
   --> gai-rs\vendor\NeuroFlow\src\data\mod.rs:194:22
    |
194 |             for j in 0..self.x[i].len() {
    |                      ^^^^^^^^^^^^^^^^^^
    |
    = note: `#[warn(clippy::needless_range_loop)]` on by default
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#needless_range_loop
help: consider using an iterator
    |
194 |             for (j, <item>) in sum_x.iter_mut().enumerate().take(self.x[i].len()) {
    |                 ~~~~~~~~~~~    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

warning: the loop variable `j` is used to index `sum_y`
   --> gai-rs\vendor\NeuroFlow\src\data\mod.rs:200:22
    |
200 |             for j in 0..self.y[i].len() {
    |                      ^^^^^^^^^^^^^^^^^^
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#needless_range_loop
help: consider using an iterator
    |
200 |             for (j, <item>) in sum_y.iter_mut().enumerate().take(self.y[i].len()) {
    |                 ~~~~~~~~~~~    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

I get the suggestion, but is it really better in this case? Is it a zero-cost abstraction?

Regardless of whether it compiles to the same code (I doubt it does, but I haven't checked), it's definitely harder to reason about.

Is there a better suggestion clippy could make for summing up all the items in a Vec<Vec<_>>?

Start with the needless_range_loop lint and the example in gai-rs\vendor\NeuroFlow\src\data\mod.rs around lines 194 and 200. Compare the current iterator suggestion with the nested Vec<Vec> summation shown in the issue, then determine what revised suggestion would be clearer and preserve the intended behavior.】【。