[Java] FieldVector getFieldBuffers API should not set reader/writer indices

Per discussion .

The fact that we have reader/writer settings in `getFieldBuffers` is wrong. To clarify, `getFieldBuffers` is distinct from `getBuffers`. The former should be for getting access to underlying data for higher-performance algorithms. The latter is for sending the data over the wire. Seems we've mixed up use of both.

Currently in `VectorUnloader`, we used `getFieldBuffers` to create `ArrowRecordBatch` that’s why we keep writer/reader indices in `getFieldBuffers`, we should use `getBuffers` instead.

**Reporter**: [Ji Liu](https://issues.apache.org/jira/browse/ARROW-7539) / @tianchen92
#### PRs and other links:
- [GitHub Pull Request apache/arrow#6156](https://github.com/apache/arrow/pull/6156)

_{**Note**: *This issue was originally created as [ARROW-7539](https://issues.apache.org/jira/browse/ARROW-7539). Please see the [migration documentation](https://github.com/apache/arrow/issues/14542) for further details.*}

Comece com VectorUnloader e compare o uso de getFieldBuffers e getBuffers ao criar instâncias de ArrowRecordBatch. Leia a discussão sobre a API FieldVector vinculada na issue e, em seguida, verifique se getFieldBuffers fornece os dados subjacentes sem alterar os índices de leitura/gravação, enquanto a serialização de record-batch usa getBuffers.

Escrita pelo modelo de indexação a partir do texto da issue.