Description of problem:

    The toolbox command always ignores [command] the first time.


Version-Release number of selected component (if applicable):

    toolbox-0.0.8-3.rhaos4.8.el8.noarch

How reproducible:


The first time:
```
sh-4.4# toolbox ls                 # <== add `ls`
Trying to pull registry.redhat.io/rhel8/support-tools:latest...
Getting image source signatures
Checking if image destination supports signatures
Copying blob adffa6963146 done  
Copying blob 29bfe07dbe66 done  
Copying blob 26f1167feaf7 done  
Copying config e0a3910607 done  
Writing manifest to image destination
Storing signatures
e0a3910607766bb10a9d84a0223db78d9856407e34af50a4f8fc0587c6c041a9
Spawning a container 'toolbox-root' with image 'registry.redhat.io/rhel8/support-tools'
Detected RUN label in the container image. Using that as the default...
[root@master-1 /]#                 # <== ls will not run
[root@master-1 /]# 
[root@master-1 /]# exit
```


From the second time onwards:
```
sh-4.4# 
sh-4.4# toolbox ls
Container 'toolbox-root' already exists. Trying to start...
(To remove the container and start with a fresh toolbox, run: sudo podman rm 'toolbox-root')
toolbox-root
Container started successfully. To exit, type 'exit'.
bin  boot  dev    etc  home  host  lib  lib64  lost+found  media    mnt  opt  proc    root  run  sbin  srv  sys  tmp    usr  var
sh-4.4# 
```


Steps to Reproduce:
    1. Execute it with the [command] option on the node that has never started toolbox.

Actual results:

    The toolbox command always ignores [command] the first time.


Expected results:

    The [command] is executed even at the first time.


Additional info:

There are no notes in the --help:

```
sh-4.4# toolbox --help
USAGE: toolbox [-h/--help] [command]
toolbox is a small script that launches a container to let you bring in your favorite debugging or admin tools.
The toolbox container is a pet container and will be restarted on following runs.
To remove the container and start fresh, do sudo podman rm toolbox-root.
```

The case where [command] is executed is the pattern that through container_exec(), but it is not executed because it through only container_runlabel() at the first time.

https://github.com/coreos/toolbox/blob/0.0.8/rhcos-toolbox

```
run() {
    if ! image_exists; then
        image_pull
    fi

    local runlabel=$(image_runlabel)
    if ! container_exists; then
        echo "Spawning a container '$TOOLBOX_NAME' with image '$TOOLBOX_IMAGE'"
        if [[ -z "$runlabel" ]] || [[ "$runlabel" == "<no value>" ]]; then
            container_run
            return
        else
            echo "Detected RUN label in the container image. Using that as the default..."
            container_runlabel                   <== here
            return
        fi
    else
        echo "Container '$TOOLBOX_NAME' already exists. Trying to start..."
        echo "(To remove the container and start with a fresh toolbox, run: sudo podman rm '$TOOLBOX_NAME')"
    fi

    local state=$(container_state)
    if [[ "$state" == configured ]] || [[ "$state" == exited ]] || [[ "$state" == stopped ]]; then
        container_start
    elif [[ "$state" != running ]]; then
        echo "Container '$TOOLBOX_NAME' in unknown state: '$state'"
        return 1
    fi

    echo "Container started successfully. To exit, type 'exit'."
    container_exec "$@"
}
```

```
container_runlabel() {
    if ! sudo podman container runlabel --name "$TOOLBOX_NAME" RUN "$TOOLBOX_IMAGE" 2>&1; then
        echo "$0: failed to runlabel on image '$TOOLBOX_IMAGE'"
        exit 1
    fi
}
```

```
container_exec() {
    local cmd=$@
    if [ ! -n "$cmd" ]; then
        cmd=$(sudo podman inspect "$TOOLBOX_IMAGE" | jq -re ".[].Config.Cmd[0]") || cmd="/bin/sh"
    fi

    sudo podman exec \
        --env LANG="$LANG" \
        --env TERM="$TERM" \
        --tty \
        --interactive \
        "$TOOLBOX_NAME" \
        "$cmd"                   <== here
}
```