[SPARK-49382][PS] Make frame box plot properly render the fliers/outliers

[zhengruifeng]

What changes were proposed in this pull request?

fliers/outliers was ignored in the initial implementation #36317

Why are the changes needed?

feature parity for Pandas and Series box plot

Does this PR introduce any user-facing change?

import pyspark.pandas as ps
df = ps.DataFrame([[5.1, 3.5, 0], [4.9, 3.0, 0], [7.0, 3.2, 1], [6.4, 3.2, 1], [5.9, 3.0, 2], [100, 200, 300]], columns=['length', 'width', 'species'])
df.boxplot()

df.length.plot.box()

before:
df.boxplot()

after:
df.boxplot()

How was this patch tested?

CI and manually check

Was this patch authored or co-authored using generative AI tooling?

No

[zhengruifeng]

I feel it weird to select the outliers by the distance |value - lower_whisker|, which is used in series.boxplot.

It should be something like |value - median| or |value - mean|, will revisit this later.

[zhengruifeng]

merged to master