Hvplot with datashade/rasterize not working neither in Google Colab nor Kaggle

Previously you were able to run hvplot with full functionality on those platforms just by installing hvplot with:
```
!pip install hvplot
```

Since the last mayor version update, to make hvplot work you have to:
```
!pip install --upgrade hvplot
!pip install --upgrade bokeh
!pip install --upgrade holoviews
```

However, I have not been able to make it work with datashader/rasterize options, even after installing:
```
!pip install datashader
!pip install jupyter_bokeh
```

Reproducible example:
```
# Load iris dataset
import pandas as pd
from sklearn.datasets import load_iris

iris = load_iris()
df = pd.DataFrame(data=iris.data, columns=iris.feature_names)
df['target'] = iris.target
```

```
# this will work:
import hvplot.pandas

hvplot.extension('bokeh')
df.hvplot(kind='scatter', x='sepal width (cm)', y='sepal length (cm)', width=600, height=300)
```

```
# this will not work:
import hvplot.pandas

hvplot.extension('bokeh')
df.hvplot(kind='scatter',
x='sepal width (cm)',
y='sepal length (cm)',
groupby='target',
width=600,
height=300,
datashade=True,
# rasterize=True,
cnorm= 'eq_hist',
dynspread=True,
alpha=0.5)
```

Any hints to make hvplot with datashader work on this online platforms would be welcome. Thanks in advance.

YbY.