Cannot create a SLURM cluster using dask-labextension

Hello,

My name is Clément Buton, I am a Post-doc working at [IPNL](https://www.ipnl.in2p3.fr/?lang=en) on the[ Euclid European space telescope](https://www.euclid-ec.org).

I am currently trying to use dask-labextension with a SLURM cluster on my laboratory grid infrastructure. So far, I have no problem using dask-labextension together with the local cluster configuration or opening a SLURM cluster client "by hand":

```
from dask.distributed import Client
from dask_jobqueue import SLURMCluster
cluster = SLURMCluster(cores=16,
project="euclid_nisp",
local_directory='$TMPDIR',
walltime='01:00:00',
memory="50GB",
queue="normal")
cluster.scale(10)
client = Client(cluster, processes=False)
client
```

My problem starts when I try to configure the dask-labextension (labextension.yaml) with the SLURMCluster as follows:

```
labextension:
factory:
module: 'dask_jobqueue'
class: 'SLURMCluster'
args: []
kwargs: {'cores': 16,
'processes': 1,
'memory': '50GB',
'queue': 'normal',
'walltime': '02:00:00',
'local_directory': '$TMPDIR'}
default:
workers: null
adapt: null
# minimum: 0
# maximum: 10
initial: []
```

I get the following error message when I try to create a new cluster (i.e. click on the +new button): "cannot get address of non-running Server" (cf a copy of my screen [jupyterlab-dask-slurm.pdf](https://github.com/dask/dask-labextension/files/3358676/jupyterlab-dask-slurm.pdf)) which seems to be related to dask.distributed.

I add bellow the versions of the libraries and extensions I am currently using:

```
(py37) ✘-1 ~/dask-slurm
11:09 $ jupyter serverextension list
config dir: /home/euclid/cbuton/.miniconda/envs/py37/etc/jupyter
dask_labextension enabled
- Validating...
dask_labextension 0.4.0a1 OK
jupyter_server_proxy enabled
- Validating...
jupyter_server_proxy OK
jupyterlab enabled
- Validating...
jupyterlab 1.0.1 OK
jupyterlab_slurm enabled
- Validating...
jupyterlab_slurm 0.1.1 OK
nbserverproxy enabled
- Validating...
nbserverproxy OK

(py37) ✔ ~/dask-slurm
11:10 $ conda list
jupyterlab 1.0.1
jupyterlab-server 1.0.0
dask 2.0.0
dask-core 2.0.0
dask-jobqueue 0.5.0
dask-labextension 0.4.0a1
```

Any help would be deeply appreciated.
Thanks in advance, and thanks again on the really nice work you guys are doing,

Sincerely,
Clément