Change dtype of ElementIdentifiers and DynamicTableRegion to int32

[rly]

Summary of changes

Be explicit about the minimum integer precision required for ElementIdentifiers and DynamicTableRegion being int32. This is in anticipation of a change in the meaning of dtype: int to mean int8 instead of int32.

If the id columns of a DynamicTable (type: ElementIdentifiers) is initialized to be an int8 dataset and is set to be expandable, then you cannot add more than 128 rows to the table. You would have to delete the dataset and re-create it with a different dtype. Similarly, if a DynamicTableRegion is initialized to be an int8 dataset, referenced row indices could not be larger than 128. This is rare and may never come up, but is why @oruebel and I would prefer to keep the dtype of ElementIdentifiers and DynamicTableRegion to be int32 instead of int8.

See

• Allow arbitrary precision for generic numeric types NeurodataWithoutBorders/pynwb#1631 (comment)
• Change "int" meaning to "int8", reorder table hdmf-schema-language#38

PR checklist for schema changes

• Update the version string in docs/source/conf.py and common/namespace.yaml to the next version with the suffix "-alpha"
• Add a new section in the release notes for the new version with the date "Upcoming"
• Add release notes for the PR to docs/source/hdmf_common_release_notes.rst and/or
  docs/source/hdmf_experimental_release_notes.rst