Return status code after testing

[AlfonZ42]

Currently, even if some tests fail, status code returned is 0, thus CI doesn't find out that there was failure and build doesn't fail. This changes it.

Current behavior:

• Non-Windows, tests pass - process finishes, status code is zero
• Non-Windows, tests fail - process finishes, status code is zero
• Windows, tests pass - process keeps running
• Windows, tests fail - process keeps running

Keeping the process running under Windows is desired behavior - see 9ad7e38.

New behavior:

• Non-Windows, tests pass - process finishes, status code is zero
• Non-Windows, tests fail - process finishes, status code is non-zero (CI reports failed build)
• Windows, tests pass - process keeps running
• Windows, tests fail - process keeps running

I have disabled verbose mode for now, even though 9ad7e38 intended to have it enabled.
Due to bug (which this change fixes) it wasn't taken into account. Enabling it would report passed tests as well, making failed tests less visible, see below.

Non-verbose output (with fake failed test added):

-- Starting suite "test/test_maps", 4 test(s)
  .F..
-- Starting suite "main", 87 test(s)
  ...................................................................
  ....................
---- Testing finished in 4.00 s, with 911 assertion(s) ----
  90 passed, 1 failed, 0 error(s), 0 skipped.
FAIL: test/test_maps.test_fakefail (0.00ms): (Failed) - Fake failed test. (286)

Verbose output:

Starting tests, 16 suite(s)
-- Starting suite "test/test_maps", 4 test(s)
PASS: test_breakable_blocks (0.00ms)
FAIL: test_fakefail (0.00ms): (Failed) - Fake failed test. (286)
PASS: test_music (140.63ms)
PASS: test_doors (15.63ms)
    Finished suite "test/test_maps", +3 -1 E0 s0
-- Starting suite "main", 87 test(s)
PASS: test_fsm_cancel_onbefore_event (0.00ms)
PASS: test_load_current (46.88ms)
PASS: test_tween_handle_anything (0.00ms)
PASS: test_get_action (0.00ms)
PASS: test_queue_push_events (0.00ms)
PASS: test_split_string (0.00ms)
PASS: test_sparkle_same_versions (0.00ms)
PASS: test_fsm_state_transition (0.00ms)
PASS: test_give_gcc_key (0.00ms)
PASS: test_transition_new (0.00ms)
PASS: test_dialog_whole_message (0.00ms)
PASS: test_sparkle_osx_unzip_unknown_file (0.00ms)
PASS: test_dialog_update_cursor (0.00ms)
PASS: test_round_down (0.00ms)
PASS: test_randomness (0.00ms)
PASS: test_find_related_costume (0.00ms)
PASS: test_fsm_fire_onwarn (0.00ms)
PASS: test_round_up (0.00ms)
PASS: test_track_row (0.00ms)
PASS: test_init_no_args (0.00ms)
PASS: test_transition_time (0.00ms)
PASS: test_find_unrelated_costume (15.63ms)
PASS: test_queue_new (0.00ms)
PASS: test_add_scroll_to_inventory_and_select (0.00ms)
PASS: test_remove_all_items (0.00ms)
PASS: test_fsm_no_yellow (0.00ms)
PASS: test_sparkle_lower_minor_version (0.00ms)
PASS: test_give_taco_meat (0.00ms)
PASS: test_distinct_id_len (15.63ms)
PASS: test_pick_unknown_costume (0.00ms)
PASS: test_sparkle_higher_minor_version (0.00ms)
PASS: test_remap_persistence (0.00ms)
PASS: test_fsm_change_handler (0.00ms)
PASS: test_transition_go_forward (0.00ms)
PASS: test_updater_create (0.00ms)
PASS: test_keyinuse (0.00ms)
PASS: test_updater_progress_not_started (0.00ms)
PASS: test_get_key (0.00ms)
PASS: test_keymap (0.00ms)
PASS: test_osx_get_application_path (0.00ms)
PASS: test_fsm_mixins (0.00ms)
PASS: test_sparkle_remove_directory (15.63ms)
PASS: test_sparkle_lower_versions (0.00ms)
PASS: test_load_unknown_character (0.00ms)
PASS: test_fsm_can (0.00ms)
PASS: test_fsm_cannot (0.00ms)
PASS: test_toggle_jump_high (0.00ms)
PASS: test_sparkle_higher_bug_versions (0.00ms)
PASS: test_give_money (0.00ms)
PASS: test_transition_go_backward (0.00ms)
PASS: test_queue_overwrite (0.00ms)
PASS: test_sparkle_unsupported_versions (0.00ms)
PASS: test_fsm_canel_warn_event (0.00ms)
PASS: test_osx_short_path (0.00ms)
PASS: test_fsm_accept_arguments (0.00ms)
PASS: test_string_endswith (0.00ms)
PASS: test_pick_unknown_character (0.00ms)
PASS: test_fsm_start_green (0.00ms)
PASS: test_remove_first_args (0.00ms)
PASS: test_get (0.00ms)
PASS: test_distinct_id (15.63ms)
PASS: test_string_startswith (0.00ms)
PASS: test_fsm_check_states (0.00ms)
PASS: test_distinct_id_source (0.00ms)
PASS: test_queue_events_arguments (0.00ms)
PASS: test_updater_no_thread_started (0.00ms)
PASS: test_sign_negvative (0.00ms)
PASS: test_pick_known_combination (0.00ms)
PASS: test_load_abed (0.00ms)
PASS: test_flush_row (15.63ms)
PASS: test_sparkle_lower_major_version (0.00ms)
PASS: test_random_id (0.00ms)
PASS: test_osx_no_root_path (0.00ms)
PASS: test_remove_duplicate_args (0.00ms)
PASS: test_sparkle_windows_basename (0.00ms)
PASS: test_sign_zero (0.00ms)
PASS: test_unlock_levels (0.00ms)
PASS: test_tween_hanlde_nil (0.00ms)
PASS: test_queue_validate_arguments (0.00ms)
PASS: test_sparkle_higher_major_version (0.00ms)
PASS: test_sign_positive (0.00ms)
PASS: test_sparkle_remove_path_no_exist (0.00ms)
PASS: test_use_respect (31.25ms)
PASS: test_dialog_message (0.00ms)
PASS: test_filename (0.00ms)
PASS: test_tween_negative_value (0.00ms)
PASS: test_tween_zero_value (0.00ms)
    Finished suite "main", +87 -0 E0 s0

---- Testing finished in 3.98 s, with 911 assertion(s) ----
  90 passed, 1 failed, 0 error(s), 0 skipped.

More importantly, there is a bug in lunatest that causes not reporting suite_setup's error in verbose mode - see silentbicycle/lunatest#14.