[SPARK-12938][SQL] DataFrame API for Bloom filter

[cloud-fan]

This PR integrates Bloom filter from spark-sketch into DataFrame. This version resorts to RDD.aggregate for building the filter. A more performant UDAF version can be built in future follow-up PRs.

This PR also add 2 specify put version(putBinary and putLong) into BloomFilter, which makes it easier to build a Bloom filter over a DataFrame.

[cloud-fan]

cc @rxin @liancheng

[SparkQA]

Test build #50156 has finished for PR 10937 at commit a0dcaa8.

• This patch fails PySpark unit tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • public class BloomFilterImpl extends BloomFilter implements Serializable

[yhuai]

Also add comment to explain the branching at here?

[rxin]

specific -> specialized

version -> variant

[rxin]

Since the two (cms and bf) are implemented by two different persons, it'd be great for one of you to go through both to make sure everything is consistent. We can do that in a follow-up pull request.

[cloud-fan]

retest this please

[SparkQA]

Test build #50208 has finished for PR 10937 at commit bd0671c.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • public class Utils

[rxin]

Thanks - going to merge this.