[Java] Implement non-sparse tensors

We'd like to be able to round-trip NumPy ndarrays through Java, and create tensors in Java that can be eventually mapped to ndarrays in Python. Having even a basic Tensor implementation, with extension types, as a contrib module would help greatly.

Some prior discussions
-

**Reporter**: [David Li](https://issues.apache.org/jira/browse/ARROW-10101) / @lidavidm
#### PRs and other links:
- [GitHub Pull Request apache/arrow#8278](https://github.com/apache/arrow/pull/8278)

_{**Note**: *This issue was originally created as [ARROW-10101](https://issues.apache.org/jira/browse/ARROW-10101). Please see the [migration documentation](https://github.com/apache/arrow/issues/14542) for further details.*}

Start by reading the linked prior discussion and Apache Arrow pull request #8278; the issue does not name Java files, tests, or entry points. Done would mean supporting non-sparse tensors that can round-trip NumPy ndarrays and be created in Java for later mapping to Python ndarrays.

Written by the indexing model from the issue text.