feat(algorithms): add IpopCmaEs (CMA-ES with restart) for multimodal problems

Auger & Hansen 2005 IPOP-CMA-ES: wraps the existing CmaEs in a restart
loop that doubles the population size and re-randomizes the mean
whenever a restart trigger fires. Specifically addresses the failure
mode we observed on Rastrigin (vanilla CMA-ES = 2.3 vs DE = 0).

Restart triggers:
- The whole budget for one inner CmaEs run finishes without improvement
- (More sophisticated triggers — eigenvalue collapse, condition-number
  blow-up, sigma stagnation — are left for future versions; the
  per-run budget trigger captures the bulk of the practical benefit)

Each restart:
- Doubles the population_size (Auger & Hansen 2005)
- Re-randomizes the initial mean to a fresh point in the bounds box
- Resets sigma to the user's initial value

Same Vec<f64> + single-objective constraints as CmaEs. The total
budget is divided across restarts; restart budget grows with
population. Tests verify it beats vanilla CMA-ES on Rastrigin.

src/algorithms/one_plus_one_es.rs

@@ -97,14 +97,13 @@ where
 
         let mut sigma = self.config.initial_sigma;
         let mut window = std::collections::VecDeque::with_capacity(self.config.adaptation_period);
-        let n = parent.len();
 
         for _ in 0..self.config.iterations {
             let normal = Normal::new(0.0, sigma).expect("Normal::new(0, sigma)");
             let mut child = parent.clone();
-            for j in 0..n {
+            for (j, x) in child.iter_mut().enumerate() {
                 let (lo, hi) = self.bounds.bounds[j];
-                child[j] = (child[j] + normal.sample(&mut rng)).clamp(lo, hi);
+                *x = (*x + normal.sample(&mut rng)).clamp(lo, hi);
             }
             let child_eval = problem.evaluate(&child);
             evaluations += 1;