Create specialized HadoopParquetReader

Setting up a specialized HadoopParquetReader allows for deprecating the Hadoop specific methods in ParquetReader as in PARQUET-2381. This could allow for an easier and more gentle transition away from the current Hadoop-coupled Parquet reader API by encouraging transitioning to this specialized class already, before completely phasing out the Hadoop methods from ParquetReader in the future.

**Reporter**: [Atour Mousavi Gourabi](https://issues.apache.org/jira/secure/ViewProfile.jspa?name=amousavigourabi) / @amousavigourabi
#### Related issues:
- [Deprecate methods relying on Hadoop classes when alternatives using more generic Parquet interfaces are available](https://github.com/apache/parquet-java/issues/2838) (blocks)

_{**Note**: *This issue was originally created as [PARQUET-2403](https://issues.apache.org/jira/browse/PARQUET-2403). Please see the [migration documentation](https://issues.apache.org/jira/browse/PARQUET-2502) for further details.*}

Start by reviewing ParquetReader's Hadoop-specific methods and the related PARQUET-2381 and issue 2838 discussions. Create the specialized HadoopParquetReader described here, with the migration documentation and future deprecation path kept in view; done means the new class provides the intended transition away from Hadoop-coupled ParquetReader methods.

Written by the indexing model from the issue text.