Co-authored-by: Claude Opus 4.7 (1M context) <[email protected]>

Bumps [urllib3](https://github.com/urllib3/urllib3) from 2.6.3 to 2.7.0.
- [Release notes](https://github.com/urllib3/urllib3/releases)
- [Changelog](https://github.com/urllib3/urllib3/blob/main/CHANGES.rst)
- [Commits](urllib3/urllib3@2.6.3...2.7.0)

---
updated-dependencies:
- dependency-name: urllib3
  dependency-version: 2.7.0
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <[email protected]>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

The string is typically shared across all variables in a dataset, so
parse once per distinct attr. Wrap the cached result in `Frozen` to
prevent accidental mutation of the shared value.

Co-Authored-By: Claude Opus 4.7 (1M context) <[email protected]>

Skip the per-key _get_all fan-out when every key is already a plain
data-variable name — the slow path would just fall through to the same
result. Decoration via get_associated_variable_names is unchanged.

Co-Authored-By: Claude Opus 4.7 (1M context) <[email protected]>

drop_bounds is a no-op unless the caller passed a scalar key, so
list-key calls (ds.cf[[name]]) never need it. Inline the lookup so
list-key paths skip the full-dataset bounds scan entirely.

Co-Authored-By: Claude Opus 4.7 (1M context) <[email protected]>