Reduce memory pressure when reading footers

I encountered OOMs reading metadata for a dataset with 500+ columns, many of them quite sparse.

We can reduce memory utilization significantly.

**Reporter**: [Dmitriy V. Ryaboy](https://issues.apache.org/jira/secure/ViewProfile.jspa?name=dvryaboy) / @dvryaboy

_{**Note**: *This issue was originally created as [PARQUET-11](https://issues.apache.org/jira/browse/PARQUET-11). Please see the [migration documentation](https://issues.apache.org/jira/browse/PARQUET-2502) for further details.*}

Start by reproducing the out-of-memory condition while reading metadata for a dataset with 500+ sparse columns, then trace the footer-reading path. The issue names no files or tests, so identify the relevant metadata and footer code before deciding on the change. Done means substantially lower memory utilization without changing the metadata read result.

Written by the indexing model from the issue text.