test/spdkcli: Wait until spdkcli_clear_config settled

Similarly to https://review.spdk.io/gerrit/c/spdk/spdk/+/17559, this
part may also be affected by:

Error in cmd: load_config /root/spdk/test/spdkcli/config.json (
load_config /root/spdk/test/spdkcli/config.json
request:
{
  "action_on_timeout": "none",
  "timeout_us": 0,
  "timeout_admin_us": 0,
  "keep_alive_timeout_ms": 10000,
  "transport_retry_count": 4,
  "arbitration_burst": 0,
  "low_priority_weight": 0,
  "medium_priority_weight": 0,
  "high_priority_weight": 0,
  "nvme_adminq_poll_period_us": 10000,
  "nvme_ioq_poll_period_us": 0,
  "io_queue_requests": 1024,
  "delay_cmd_submit": true,
  "bdev_retry_count": 3,
  "transport_ack_timeout": 0,
  "ctrlr_loss_timeout_sec": 0,
  "reconnect_delay_sec": 0,
  "fast_io_fail_timeout_sec": 0,
  "generate_uuids": false,
  "transport_tos": 0,
  "io_path_stat": false,
  "method": "bdev_nvme_set_options",
  "req_id": 31
}
Got JSON-RPC error response
response:
{
  "code": -1,
  "message": "RPC not permitted with nvme controllers already attached"
}

so make sure the nvme controller was fully detached.

Change-Id: Iaebed7b640d96fbabfc694dfebc5a725902caad2
Signed-off-by: Michal Berger <michal.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/17850
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>

test/spdkcli/vhost.sh

@@ -122,6 +122,10 @@ $spdkcli_job "'vhost/scsi/vhost_scsi2 remove_target 2' 'Nvme0n1p3'
 "
 timing_exit spdkcli_clear_config
 
+# Make sure we wait long enough for the nvme ctrl to disappear, otherwise
+# we are at risk of hitting -EPERM while loading bdev configuration.
+xtrace_disable_per_cmd wait_for_all_nvme_ctrls_to_detach
+
 timing_enter spdkcli_load_config
 $spdkcli_job "'load_config $testdir/config.json'
 '/lvol_stores create lvs0 Malloc0' 'lvs0' True