DataChunkIterator clarify iteration behavior

[oruebel]

Description

DataChunkIterator currently assumes that the iterator yields one element along the iteration dimensions at a time. We should clarify this in the documentation:

• Add to the main docstring of DataChunkIterator a note that describes the expected iteration behavior
• Add note to DataChunkIterator.maxshape to indicate that for iterators this adds a dimension to the chunk generated by the iterator. E.g. if the iterator produces chunks of (x,y) the DataChunkIterator will have a shape of (None, x, y)
• Add a corresponding note also to the iterative write tutorial https://pynwb.readthedocs.io/en/stable/tutorials/general/iterative_write.html#iterative-data-write

Alternative approach:

• Enhance DataChunkIterator to allow iterators that return multiple blocks of data along the iteration dimension. This is useful in case where, e.g., reading single timesteps from a file is very expensive and we want the iterator to return a range of timesteps directly from the file. This would require an additional parameter to control this behavior, e.g., multistep_chunks=True to indicate that chunks retrieved from the iterator contain multiple parts along the iteration dimension.

This has come up on the NWB Slack channel https://nwb-users.slack.com/archives/C5XKC14L9/p1622668229009900?thread_ts=1622588105.009100&cid=C5XKC14L9

Checklist

• Have you ensured the bug was not already reported ?
• Have you included a brief and descriptive title?
• Have you included a clear description of the problem you are trying to solve?
• Have you included a minimal code snippet that reproduces the issue you are encountering?
• Have you checked our Contributing document?