refactor: simplify sub_matrix with iterator-driven column population

Refactored `sub_matrix` to delegate column population to `fill_sub_matrix`, replacing manual buffer allocation and explicit permutation loops with direct value assignment via `enumerate_slots_values`. This eliminates intermediate allocations, reduces pipeline overhead, and simplifies control flow while preserving slot ordering and permutation semantics.

src/obicompactvec/src/intmatrix.rs

@@ -368,13 +368,8 @@ impl PersistentCompactIntMatrix {
     /// in the same order as `slots`.  Column access is sequential to maximize
     /// cache efficiency on the underlying mmap.
     pub fn sub_matrix(&self, slots: &[usize]) -> Vec<Vec<u32>> {
-        let n_cols = self.n_cols();
+        let mut out: Vec<Vec<u32>> = (0..self.n_cols()).map(|_| Vec::new()).collect();
-        let mut out: Vec<Vec<u32>> = Vec::with_capacity(n_cols);
+        self.fill_sub_matrix(slots, &mut out);
-        for c in 0..n_cols {
-            let mut col_buf = vec![0u32; slots.len()];
-            self.col_view(c).fill_slots_values(slots, &mut col_buf);
-            out.push(col_buf);
-        }
         out
     }
 
@@ -396,10 +391,8 @@ impl PersistentCompactIntMatrix {
         let sorted_slots: Vec<usize> = perm.iter().map(|&i| slots[i]).collect();
         for (c, col) in out.iter_mut().enumerate() {
             col.resize(n, 0);
-            let mut tmp = vec![0u32; n];
+            for (i, v) in self.col_view(c).enumerate_slots_values(&sorted_slots) {
-            self.col_view(c).fill_slots_values_sorted(&sorted_slots, &mut tmp);
+                col[perm[i]] = v;
-            for (i, &orig_idx) in perm.iter().enumerate() {
-                col[orig_idx] = tmp[i];
             }
         }
     }