Add NaNs to Data Generators In Floating-Point Testing [databricks]

[razajafri]

This PR addresses an issue where our integration tests were not considering NaNs as part of floating-point values when testing various operators. NaN (Not-a-Number) is a critical value in floating-point arithmetic, and our tests should accurately reflect its behavior. We are addressing by removing *no_nans data generators in test files unless where they are necessary. e.g. versions of Spark that don't support NaN equality like in array_test.py

Changes Made:

• Removed data generators that were excluding NaNs from floating-point values in test cases.
• Updated affected test cases to include NaNs and ensure proper coverage for various operators.
• Also renamed some of the variables to make a little bit more sense after this change.

Testing:

Ran the entire integration_test suite and saw no failures

fixes #8951

[kuhushukla]

@razajafri Add signoff but given the context I have, changes lgtm

[razajafri]

build

[razajafri]

Thank you @jlowe for the review and merge