TestParquetReader fails with hadoop 3.4.2

### Describe the bug, including details regarding any error messages, version, and platform.

A couple of test cases of `TestParquetReader` suite started failing with errors like:
```
- testRangeFiltering[0] *** FAILED ***
org.apache.parquet.bytes.TrackingByteBufferAllocator$LeakedByteBufferException: 24 ByteBuffer object(s) is/are remained unreleased after closing this allocator.
at org.apache.parquet.bytes.TrackingByteBufferAllocator.close(TrackingByteBufferAllocator.java:161)
at org.apache.parquet.hadoop.TestParquetReader.closeAllocator(TestParquetReader.java:175)
at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:77)
at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.base/java.lang.reflect.Method.invoke(Method.java:569)
```
with hadoop 3.4.2 and parquet-mr 1.16.0
The leaks are happening when reading using vectored IO because we never pass the buffers to the releaser.

### Component(s)

_No response_

Comienza con los casos fallidos de TestParquetReader, especialmente testRangeFiltering, e inspecciona la ruta de lectura de vectored IO. Comprueba cómo gestiona el releaser los búferes devueltos y vuelve a ejecutar la suite con Hadoop 3.4.2 y parquet-mr 1.16.0. Se considera terminado cuando las pruebas pasan sin errores de fugas de TrackingByteBufferAllocator.

Escrito por el modelo de indexación a partir del texto del issue.