clicktoshow not working for annotations on log scale

The click event does not seem to be triggered when clicking on an annotated point (even after waiting for the hover text to show up indicating that the point "has been found"). In the example below, the annotation cannot be toggled. With linear axes scaling (dropping the log_x|y and not rescaling x|y=np.log10...) the annotation toggling works as expected.

This bug is probably related to the (forced / unnatural) requirement to rescale the (x,y) coordinates of annotations (see #2580)

x = np.arange(1,5)
df = pd.DataFrame(data={'x':x,'y':x**2,'z':x**3,})

fig = px.line(df,
              x='x',
              y=['y','z'],
              log_x=True,
              log_y=True,
              )

annotations = []
fig.for_each_trace(
    lambda trace: annotations.append(dict(x=np.log10(trace.x[-1]),
                                          y=np.log10(trace.y[-1]),
                                          text=f'({trace.x[-1]:g},{trace.y[-1]:g})',
                                          clicktoshow='onoff',
                                          visible=True,
                                          opacity=0.5,
                                          ))
)

fig.update_layout(annotations=annotations,
                  width=300,
                  height=300,
                  )
fig.show()

首先运行提供的带对数坐标轴的 Python 示例，并将其与线性坐标轴版本进行比较。跟踪注释坐标如何重新缩放，以及 clicktoshow 如何处理对已注释点的点击，同时包括 #2580 中的相关上下文。完成标准是：在对数缩放的图表中点击时注释会切换，且不会导致线性缩放回归。