Fixing doc build conda solve and missing matplotlib output (#1009)

* Use py3.8 for conda build env
* Updated recommendation for %matplotlib widget after using inline

.github/workflows/build.yaml

@@ -30,6 +30,7 @@ jobs:
       - uses: conda-incubator/setup-miniconda@v2
         with:
           miniconda-version: "latest"
+          python-version: 3.8
       - name: Fetch unshallow
         run: git fetch --prune --tags --unshallow -f
       - name: Set output

.github/workflows/docs.yaml

@@ -40,7 +40,7 @@ jobs:
           conda config --set always_yes True
           conda install -c pyviz "pyctdev>=0.5"
           doit ecosystem_setup
-          doit env_create ${{ env.CHANS_DEV}} --python=3.7
+          doit env_create ${{ env.CHANS_DEV}} --python=3.8
       - name: doit develop_install
         run: |
           eval "$(conda shell.bash hook)"

doc/conf.py

@@ -4,7 +4,7 @@
 
 project = u'Datashader'
 authors = u'Datashader contributors'
-copyright = u'2016-2019, Anaconda, Inc.'
+copyright = u'2016-2021, Anaconda, Inc.'
 description = 'Turns even the largest data into images, accurately.'
 
 from datashader import __version__
@@ -44,7 +44,7 @@
     'SOCIAL': (
         ('Github', 'https://github.com/holoviz/datashader/'),
         ('Twitter', 'https://twitter.com/datashader/'),
-        ('Gitter', 'https://gitter.im/pyviz/pyviz'),
+        ('Discourse', 'https://discourse.holoviz.org/'),
         ('HoloViz', 'https://holoviz.org'),
     )
 })

examples/getting_started/3_Interactivity.ipynb

@@ -289,9 +289,16 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "# Choose the interactive mode for Jupyter Notebook or Lab if desired\n",
-    "#%matplotlib notebook\n",
-    "#%matplotlib widget"
+    "# Choose the interactive 'widget' mode for Jupyter Notebook or Lab if desired (instead of inline used on the website)\n",
+    "#%matplotlib widget\n",
+    "%matplotlib inline"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Note that if you wish to try the interactivity offered by `%matplotlib widget` you will need to pip or conda install the `ipympl` package."
    ]
   },
   {
@@ -397,7 +404,8 @@
     "artist3 = dsshow(df, ds.Point('x', 'y'), norm='log', cmap=alpha_colormap('#ffffff', 40, 255), ax=grid[3])\n",
     "grid[3].set_facecolor('#000000');\n",
     "grid.cbar_axes[3].set_visible(False);\n",
-    "grid[3].set_title('Point density (log, alpha-based colormap)');"
+    "grid[3].set_title('Point density (log, alpha-based colormap)');\n",
+    "fig"
    ]
   },
   {