Check for None result in gRPC (#3380)

[kevmodo]

Description

This is a fix for #3380 that checks explicitly for a None result rather than evaluating truthiness. PubSub results with zero messages evaluate to False, causing span.end() to be called twice.

Fixes #3380

Type of change

• Bug fix (non-breaking change which fixes an issue)

How Has This Been Tested?

• Ran Tox
• Ran locally against an empty PubSub queue and no longer received warning messages, but spans correctly started and ended.
• Ran against queue with messages and behavior was as expected (spans correctly started and ended, as before)

Does This PR Require a Core Repo Change?

• Yes. - Link to PR:
• No.

Checklist:

See contributing.md for styleguide, changelog guidelines, and more.

• Followed the style guidelines of this project
• Changelogs have been updated
• Unit tests have been added
• Documentation has been updated

[linux-foundation-easycla]

The committers listed above are authorized under a signed CLA.

• ✅ login: kevmodo / name: Kevin Rauwolf (0282968, 84a33db, ccf0af7, 3219568, edc2ec6, ee67c46, 3e8714b)

[xrmx]

Thanks for the PR, could you please write a test for this to avoid regressions in the future?

[kevmodo]

Thanks for the PR, could you please write a test for this to avoid regressions in the future?

That was an adventure! Added a test that counts span.end() invocations.

[xrmx]

With the small nit fixed, LGTM