[BUG] Wedge/AnnularWedge hit testing cuts off top/bottom

[mbrown1413]

Software versions

Tested on both 2.2.3 and the latest "branch-3.0" development branch.

Python version : 3.8.13 | packaged by conda-forge | (default, Mar 25 2022, 06:04:10)
IPython version : 8.2.0
Tornado version : 6.1
Bokeh version : 3.0.0dev5+23.g6f1e39a2a.dirty
BokehJS static path : /home/msbrown/bkdev/bokeh/bokeh/server/static
node.js version : v16.14.2
npm version : 8.5.0
Operating system : Linux-5.13.0-27-generic-x86_64-with-glibc2.10

Browser name and version

No response

Jupyter notebook / Jupyter Lab version

No response

Expected behavior

The Wedge and AnnularWedge glyphs should consider the whole height of the circle for hit testing, including the top and bottom.

Observed behavior

Depending on the X and Y scale, the tops and bottoms of these glyphs are cut off for hit-testing purposes. The user cannot click to select or hover over these parts.

The amount that is cut off depends on the difference in scale between the X and Y axes. This is because these glyphs are drawn as parts of circles in screen space, not in data space. Because of this, the rectangle given to the index (this.index.indices({x0, x1, y0, y1})) is incorrect.

Example code

See the [pie chart example](https://docs.bokeh.org/en/latest/docs/gallery/pie_chart.html) in the gallery.

Stack traceback or browser console output

No response

Screenshots

[mbrown1413]

I'm working on a pull request for this.

[mbrown1413]

Code for reproducing this with AnnularWedge as well:

from math import pi

import pandas as pd

from bokeh.palettes import Category20c
from bokeh.plotting import figure, show
from bokeh.transform import cumsum

x = {
    'United States': 157,
    'United Kingdom': 93,
    'Japan': 89,
    'China': 63,
    'Germany': 44,
    'India': 42,
    'Italy': 40,
    'Australia': 35,
    'Brazil': 32,
    'France': 31,
    'Taiwan': 31,
    'Spain': 29
}

data = pd.Series(x).reset_index(name='value').rename(columns={'index': 'country'})
data['angle'] = data['value']/data['value'].sum() * 2*pi
data['color'] = Category20c[len(x)]

p = figure(height=350, title="Pie Chart", toolbar_location=None,
           tools="hover", tooltips="@country: @value", x_range=(-0.5, 1.0))

p.annular_wedge(x=0, y=1, outer_radius=0.4, inner_radius=0.2,
        start_angle=cumsum('angle', include_zero=True), end_angle=cumsum('angle'),
        line_color="white", fill_color='color', legend_field='country', source=data)

p.axis.axis_label = None
p.axis.visible = False
p.grid.grid_line_color = None

show(p)