Some kwarg names not allowed in summary reduction

Some `kwarg` names are not allowed in a `summary` reduction and produce an `xarray` error when trying to convert to a `Dataset`. Example:

```python
import datashader as ds
import numpy as np
import pandas as pd

x = np.arange(2)
df = pd.DataFrame(dict(
y_from = [0.0, 1.0, 0.0, 1.0, 0.0],
y_to = [0.0, 1.0, 1.0, 0.0, 0.5],
))

canvas = ds.Canvas(plot_height=7, plot_width=7)
agg = canvas.line(source=df, x=x, y=["y_from", "y_to"], axis=1, agg=ds.summary(x=ds.count()))
```
Error produced is
```python
xarray.core.variable.MissingDimensionsError: 'x' has more than 1-dimension and the same name as one of its dimensions ('y', 'x'). xarray disallows such variables because they conflict with the coordinates used to label dimensions.
```

We should check before the calculations are performed if there is a `kwarg` name that is problematic and give a better error message. Invalid names are `x` and `y`, also `n` if using a `*_n` reduction, also the name of any categorical column used in a `by` reduction.

Start with the Canvas.line call and summary reduction in the issue's example, then trace where keyword names are handled before conversion to a Dataset. Reproduce the example and verify that reserved names such as x, y, n, and categorical by-column names receive a clear validation error before calculations run.

Written by the indexing model from the issue text.