cm-runtime: don't drain the warm pool on a flaky sandbox health check

`driver.health()` can return `Err(_)` on transient docker daemon hiccups
(connection reset, timeout mid-inspect, daemon busy). exec() used
`.unwrap_or(false)` which treated Err as "dead" and:
  1. Destroyed the agent's assigned sandbox.
  2. Fell through to the warm pool and popped one, draining it by 1.
  3. Provisioned a fresh assigned sandbox on top.

Under CI load — where several test processes hit the docker daemon
concurrently — this fired as the warm_pool.rs:72 "reuse must not drain
the pool" flake. The test asserts pool_size == 2 after a reuse; when
health flaked, the reuse turned into a drain-and-refill and the assert
raced the warmer.

Match only a CONFIRMED `Ok(false)` (container dead or 404). On Err,
assume alive; if it really is dead, the subsequent exec surfaces the
error with a clear message instead of silent sandbox destruction and
warm-pool drainage.

crates/cm-runtime/src/sandboxes.rs

@@ -193,7 +193,15 @@ impl SandboxManager {
                 id: row.container_id,
                 name: row.name,
             };
-            if driver.health(&handle).await.unwrap_or(false) {
+            // Only treat the sandbox as dead on a CONFIRMED `Ok(false)`. Docker
+            // health checks flake under load (transient connection resets,
+            // daemon busy) and `.unwrap_or(false)` used to bulldoze the
+            // assigned sandbox AND drain a warm pool slot on those flakes —
+            // observed as the warm_pool.rs:72 "reuse must not drain the pool"
+            // flake. On ambiguous state (Err), assume alive; the subsequent
+            // exec surfaces any real failure with a clear error to the caller.
+            let confirmed_dead = matches!(driver.health(&handle).await, Ok(false));
+            if !confirmed_dead {
                 return driver
                     .exec(&handle, &["sh", "-lc", command])
                     .await