Add an indexable variant of Arrow.Stream

In distributed computing context it would be nice to have a vector-variant of `Arrow.Stream` iterator. The idea is to be able to split processing of a single large arrow file with multiple record batches into multiple worker processes. Looking at the source code this should be possible to be done in a relatively efficient way.

@quinnj - what do you think?

Start by reading the existing Arrow.Stream implementation and its source code to understand how record batches are exposed. Define the indexable vector variant and its worker-splitting behavior before implementation; done means a large Arrow file can be partitioned across multiple worker processes with appropriate coverage for the new API.

Written by the indexing model from the issue text.