fix(deploy): include DEGRADED deploys as restorable checkpoints

Snapshot is written by DeploymentExecutor before the RUNNING/DEGRADED
split, so DEGRADED rows already carry a deployed_config_snapshot. Treat
them as checkpoints — partial-healthy deploys still produced a working
config worth restoring. Aligns repo query with UI filter.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

cameleer-server-app/src/main/java/com/cameleer/server/app/storage/PostgresDeploymentRepository.java

@@ -140,10 +140,12 @@ public class PostgresDeploymentRepository implements DeploymentRepository {
     }
 
     public Optional<Deployment> findLatestSuccessfulByAppAndEnv(UUID appId, UUID envId) {
+        // DEGRADED deploys also carry a snapshot (executor writes before the RUNNING/DEGRADED
+        // split), and represent a config that reached COMPLETE stage — restorable for the user.
         var results = jdbc.query(
                 "SELECT " + SELECT_COLS + " FROM deployments "
                 + "WHERE app_id = ? AND environment_id = ? "
-                + "AND status = 'RUNNING' AND deployed_config_snapshot IS NOT NULL "
+                + "AND status IN ('RUNNING', 'DEGRADED') AND deployed_config_snapshot IS NOT NULL "
                 + "ORDER BY deployed_at DESC NULLS LAST LIMIT 1",
                 (rs, rowNum) -> mapRow(rs), appId, envId);
         return results.isEmpty() ? Optional.empty() : Optional.of(results.get(0));