[C++][Parquet] ArrowInvalid: Integer value emitted when reading parquet dataset

I've created a parquet db with an uint64 datatype. When reading some of the files are raising the errors like
> ArrowInvalid: Integer value 12120467241726599441 not in range: 0 to 9223372036854775807

**Environment**: Ubuntu 20.04, PyArrow 6.01, Python 3.9
**Reporter**: [mondonomo](https://issues.apache.org/jira/browse/ARROW-15146)

_{**Note**: *This issue was originally created as [ARROW-15146](https://issues.apache.org/jira/browse/ARROW-15146). Please see the [migration documentation](https://github.com/apache/arrow/issues/14542) for further details.*}

No source files, tests, or entry points are identified in the report. Reproduce the failure with the stated Ubuntu, PyArrow, and Python environment using a Parquet dataset containing uint64 values, then trace the dataset reader and verify that values outside the signed 64-bit range are read without ArrowInvalid.

Written by the indexing model from the issue text.