fix: avoid NLTK download race condition in parallel test workers

[bsbodden]

Try loading stopwords before downloading to prevent concurrent pytest-xdist workers from racing on nltk.download(), which raises FileExistsError when multiple processes write to the same path.

[Copilot]

Pull request overview

This PR fixes a race condition that occurs when multiple pytest-xdist workers attempt to download NLTK stopwords data concurrently. The fix implements a "try-load-first" pattern that checks if the data is already available before attempting to download it, preventing FileExistsError exceptions when multiple processes write to the same path simultaneously.

Changes:

• Wrapped NLTK stopwords loading in a try-except block to check if data exists before downloading
• Applied the fix consistently in both _get_stopwords and _set_stopwords methods

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
redisvl/utils/full_text_query_helper.py	Updated _get_stopwords to try loading stopwords before downloading, preventing race conditions in parallel test workers
redisvl/query/query.py	Updated _set_stopwords in TextQuery class with the same race condition fix

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[vishal-bala]

LGTM! I'd guess there's technically still a race condition in that

• two workers could try to start downloading at the same time, or
• a worker could try to load a partially downloaded dataset, fail, and then decide to start downloading it as well

But these are both fairly niche and unlikely, and probably not worth designing for.