[SPARK-30351][ML][PySpark] BisectingKMeans support instance weighting

[huaxingao]

What changes were proposed in this pull request?

add weight support in BisectingKMeans

Why are the changes needed?

BisectingKMeans should support instance weighting

Does this PR introduce any user-facing change?

Yes. BisectingKMeans.setWeight

How was this patch tested?

Unit test

[SparkQA]

Test build #115895 has finished for PR 27035 at commit 5e7ccdb.

• This patch fails Spark unit tests.
• This patch merges cleanly.
• This patch adds no public classes.

[SparkQA]

Test build #115898 has finished for PR 27035 at commit e8bd33f.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[huaxingao]

@zhengruifeng @srowen

[zhengruifeng]

nit, remove space before first i

[zhengruifeng]

Or just keep existing line?

[huaxingao]

Do you mean val d = input.map(_._1.size).first()?

[zhengruifeng]

It seems that var dataVectorWithNorm is only used to get norms, so what about removing it and make val norms = input.map(d => Vectors.norm(d._1, 2.0))...

[huaxingao]

I saw the conflict file. I will fix this after #27052 is merged, so I don't have to rebase twice.

[SparkQA]

Test build #115962 has finished for PR 27035 at commit 987ba2e.

• This patch passes all tests.
• This patch does not merge cleanly.
• This patch adds no public classes.

[zhengruifeng]

ping @huaxingao #27052 is merged

[SparkQA]

Test build #116128 has finished for PR 27035 at commit ae7948b.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[SparkQA]

Test build #116255 has finished for PR 27035 at commit 6ca078e.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[srowen]

Merged to master

[huaxingao]

Thanks! @srowen @zhengruifeng