gh-135110: Fix misleading generator.close() documentation

[cdenihan]

This PR fixes the misleading documentation for generator.close() that incorrectly stated the method 'raises a GeneratorExit' to the caller.

Problem

The current documentation in both Doc/reference/expressions.rst and Doc/howto/functional.rst states that generator.close() 'raises a GeneratorExit' which is misleading because:

• The method sends the GeneratorExit exception to the generator internally
• The method returns None to the caller
• The method does not raise the exception to the caller

Solution

Changed 'raises' to 'sends' in both documentation files for accuracy and consistency:

1. Doc/reference/expressions.rst (line 628-629):

   • Before: 'Raises a GeneratorExit at the point where the generator function was paused'
   • After: 'Sends a GeneratorExit exception to the generator at the point where the generator function was paused'

2. Doc/howto/functional.rst (line 605):

   • Before: 'raises a GeneratorExit exception inside the'
   • After: 'sends a GeneratorExit exception inside the'

Testing

The behavior described in the fixed documentation can be verified by the examples in the original issue report, which show that generator.close() always returns None.

Fixes #135110

---

📚 Documentation preview 📚: https://cpython-previews--135152.org.readthedocs.build/

• Issue: generator.close() never raises GeneratorExit #135110

[python-cla-bot]

All commit authors signed the Contributor License Agreement.

[ZeroIntensity]

I'm a little worried about the use of the word "send" here, especially considering generator.send() is a thing. I think "raises" is fine, we should just make it clearer that the actual yield expressions are what raise. Maybe we could say that close() is equivalent to throw(GeneratorExit)?

[ZeroIntensity]

It looks like the CLA got screwed up.

[cdenihan]

Just signed it again, should be all good now.

[ZeroIntensity]

Thanks!