Error while `groupby` agg must use 2D or 3D coordinates

If there is only one unique category, datashader raises and shows a blank map:
```python
File "/Users/ahuang/miniconda3/envs/hvplot/lib/python3.10/site-packages/holoviews/util/__init__.py", line 1013, in _process
return self.p.operation.process_element(element, key, **kwargs)
File "/Users/ahuang/miniconda3/envs/hvplot/lib/python3.10/site-packages/holoviews/core/operation.py", line 194, in process_element
return self._apply(element, key)
File "/Users/ahuang/miniconda3/envs/hvplot/lib/python3.10/site-packages/holoviews/core/operation.py", line 141, in _apply
ret = self._process(element, key)
File "/Users/ahuang/miniconda3/envs/hvplot/lib/python3.10/site-packages/holoviews/operation/datashader.py", line 1543, in _process
shaded = shade._process(self, agg, key)
File "/Users/ahuang/miniconda3/envs/hvplot/lib/python3.10/site-packages/holoviews/operation/datashader.py", line 1354, in _process
img = tf.shade(array, **shade_opts)
File "/Users/ahuang/miniconda3/envs/hvplot/lib/python3.10/site-packages/datashader/transfer_functions/__init__.py", line 717, in shade
raise ValueError("agg must use 2D or 3D coordinates")
ValueError: agg must use 2D or 3D coordinates
```
This is not ideal for `groupby` because it shows a blank map instead of just showing the single category.

To reproduce:
Data download: https://coast.noaa.gov/htdata/CMSP/AISDataHandler/2020/index.html

```python
import pandas as pd
import hvplot.pandas
import geoviews as gv
import pandas as pd
import datashader as ds
import colorcet as cc
import holoviews as hv
import holoviews.operation.datashader as hd
from holoviews.util.transform import lon_lat_to_easting_northing

df = pd.read_csv("AIS_2020_01_01.csv", parse_dates=True, index_col="BaseDateTime")
df = df.assign(
**pd.concat(lon_lat_to_easting_northing(df["LON"], df["LAT"]), axis=1).rename(
{"LON": "EASTING", "LAT": "NORTHING"}, axis=1
)
)

df.index = df.index.round("1Min")
df = df.sort_index()
df.index = df.index.astype(str)

vessel_types = pd.read_csv("AIS_categories.csv")

categories = {
r.num: r.category if r.category in [0, 2, 3, 6, 7, 16, 14, 19, 12, 18] else 21
for i, r in vessel_types.iterrows()
}

def category_desc(val):
"""Return description for the category with the indicated integer value"""
return vessel_types[vessel_types.category == val].iloc[0].category_desc

groups = {categories[i]: category_desc(categories[i]) for i in categories.keys()}
colors = cc.glasbey_bw_minc_20_minl_30
color_key = {
list(groups.keys())[i]: tuple(int(e * 255.0) for e in v)
for i, v in enumerate(colors[: (len(groups))][::-1])
}
legend = hv.NdOverlay(
{
groups[k]: hv.Points([0, 0], label=str(groups[k])).opts(
color=cc.rgb_to_hex(*v), size=0
)
for k, v in color_key.items()
}
)

df = df.loc[df["VesselType"].isin(color_key)]

pts = hv.Dataset(df, kdims=["EASTING", "NORTHING"], vdims=["VesselType"]).to(
hv.Points, groupby=["BaseDateTime"]
)
points = hd.dynspread(
hd.datashade(pts, aggregator=ds.count_cat("VesselType"), color_key=color_key)
)

tiles = (
hv.element.tiles.ESRI()
.opts(alpha=0.4, bgcolor="black")
.opts(responsive=True, min_height=600)
)
labels = hv.element.tiles.CartoDark().opts(alpha=0.7, level="glyph")
display(
tiles
* labels
* points.opts(show_legend=False)
* legend.opts(xaxis="bare", yaxis="bare", title="")
)
```

Works fine when it's not on the one category time

Reproduce the failure with the provided AIS example and follow the traceback from holoviews/operation/datashader.py into datashader/transfer_functions/__init__.py at shade. Check the single-category groupby path and verify that it renders the category instead of raising a coordinate-dimension error or showing a blank map.

Written by the indexing model from the issue text.