Setting custom formatter in so.Nominal().label() seems to do nothing

[rbergm]

When using the so.Nominal() scale of the objects interface, calling label() with a custom formatter seems to have no effect. Could also be that I am using the function in an incorrect way, since the API and documentation on the nominal scale are still being worked on.

Demo:

import matplotlib as mpl
import numpy as np
import pandas as pd
import seaborn.objects as so

xs = range(10)
ys = np.random.random(10)
df = pd.DataFrame(dict(x=xs, y=ys))
formatter = mpl.ticker.FuncFormatter(lambda x, pos: str(x) if x % 3 == 0 else "")

(so.Plot(df, x="x", y="y")
 .add(so.Dots())
 .scale(x=so.Nominal().label(formatter))
)

The x-axis ticks are correctly transformed to a nominal scale. However, I would expect to only see tick labels that divide by 3 based on the formatter.

Tested on Seaborn v0.13.2 and Matplotlib v3.9.0

[heimyoo]

Can I take it??