test: give the 10k-iteration sort-key property test its own timeout

Under parallel CI load the test repeatedly exceeded the default 5000 ms
per-test timeout (run 685 on main, run 699 on an unrelated PR); the
identical test passed on rerun. Locally it finishes in about 1.3 s, so
30 s is generous headroom, not a mask for a regression.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017aviRTgWCcAHUh1SBoxf6P

apps/api/src/pages/sort-key.test.ts

@@ -26,8 +26,15 @@ describe('sort-key helpers (issue #45)', () => {
    * pattern — repeatedly drop the last page between the first two — must never
    * collide and never overflow the key length, because the caller rebalances
    * when {@link nextKeyOrRebalance} returns null.
+   *
+   * Under parallel CI load the 10.000 iterations have repeatedly exceeded the
+   * default 5 s per-test timeout (runs 685, 699 — same code passed on rerun),
+   * so this test carries its own budget.
    */
-  it('10.000 adversarial reorders never collide or overflow (rebalance verified)', () => {
+  it(
+    '10.000 adversarial reorders never collide or overflow (rebalance verified)',
+    { timeout: 30_000 },
+    () => {
       // Start with five pages in a fixed order.
       let order = evenlySpacedKeys(5).map((key, i) => ({ id: `p${i}`, key }));
       let rebalances = 0;
@@ -69,5 +76,6 @@ describe('sort-key helpers (issue #45)', () => {
 
       // The adversarial pattern must have forced at least one rebalance.
       expect(rebalances).toBeGreaterThan(0);
-  });
+    },
+  );
 });