fn rows_converged(
a: &CsrMatrix<f64>,
x: &[DoubleDouble],
b: &[f64],
r: &[DoubleDouble],
tol: f64,
) -> boolExpand description
Step P2 (c): |RN(r_i)| ≤ t_i in every row, with t_i rounded against
convergence (the chapter, “How step P2 (c) rounds”).