Many classes in parquet-hadoop belong in parquet-column

There are a handful of classes, like BlockMetaData and some of the other \*MetaData classes that are currentlyin parquet hadoop but aren't hadoop specific. Which force some other classes (like the row group filter) to also live in parquet-hadoop when they really belong in parquet-column

**Reporter**: [Alex Levenson](https://issues.apache.org/jira/secure/ViewProfile.jspa?name=alexlevenson) / @isnotinvain

_{**Note**: *This issue was originally created as [PARQUET-38](https://issues.apache.org/jira/browse/PARQUET-38). Please see the [migration documentation](https://issues.apache.org/jira/browse/PARQUET-2502) for further details.*}

Start by locating BlockMetaData and the other *MetaData classes in parquet-hadoop, then inspect the row group filter and its module dependencies. The refactor is done when Hadoop-independent metadata and related filtering classes can live in parquet-column without leaving required dependencies behind.

Written by the indexing model from the issue text.