hil: ota: adjust delays to reduce false negative

When testing reason and state codes, propagation delay from the device to
the server were causing erroneous test failures. This increases the delay
before checking the updated server value, and also increases the device
delay between sending reason/state updates. The marginal difference is 2s
per reason code, or 20 seconds total increase in test time.

Signed-off-by: Mike Szczys <mike@golioth.io>

tests/hil/tests/ota/test.c

@@ -95,7 +95,7 @@ static void test_reason_and_state(void)
         {
             GLTH_LOGE(TAG, "Unable to report ota state: %d", status);
         }
-        golioth_sys_msleep(5000);
+        golioth_sys_msleep(7000);
     }
 
     GLTH_LOGI(TAG, "golioth_ota_get_state: %d", golioth_ota_get_state());

tests/hil/tests/ota/test_ota.py

@@ -178,8 +178,8 @@ async def test_reason_and_state(board, device, project, releases):
     # Test reason and state code updates
 
     for i, r in enumerate(golioth_ota_reason):
-        board.wait_for_regex_in_line("Updating status", timeout_s=12)
-        time.sleep(1.5) # Wait for server to propagate
+        board.wait_for_regex_in_line("Updating status", timeout_s=20)
+        time.sleep(3) # Wait for server to propagate
 
         await device.refresh()