destroy subsequent command buffers if we encountered an error at submission

This gets the `wgpu_test::ray_tracing::as_build::out_of_order_as_build` test to pass.

This seems to be an issue even on trunk, looking at the nr of calls to `create_command_encoder` & `destroy_command_encoder` in hal, they are not equal. So, I'm not sure why the validation layers don't raise the `VUID-vkDestroyDevice-device-05137`.

There is still an issue with previous command buffers being leaked but I will fix this in a follow-up.

Author: teoxoy

wgpu-core/src/device/queue.rs

@@ -1209,6 +1209,13 @@ impl Queue {
                             }
                         }
 
+                        if first_error.is_some() {
+                            if let Ok(cmd_buf_data) = cmd_buf_data {
+                                cmd_buf_data.destroy(&command_buffer.device);
+                            }
+                            continue;
+                        }
+
                         let mut baked = match cmd_buf_data {
                             Ok(cmd_buf_data) => {
                                 let res = validate_command_buffer(
@@ -1232,10 +1239,6 @@ impl Queue {
                             }
                         };
 
-                        if first_error.is_some() {
-                            continue;
-                        }
-
                         // execute resource transitions
                         if let Err(e) = unsafe {
                             baked.encoder.begin_encoding(hal_label(