#8843 Clean up request data references

[bn-dignitas]

Fix for #8843 Clear Request cancelFunction and deferred references to free up unintentional retained request data memory

[cesium-concierge]

Thanks for the pull request @bn-dignitas!

• ✔️ Signed CLA found.
• ❔ CHANGES.md was not updated.
  • If this change updates the public API in any way, please add a bullet point to CHANGES.md.
• ❔ Unit tests were not updated.
  • Make sure you've updated tests to reflect your changes, added tests for any new code, and ran the code coverage tool.

Reviewers, don't forget to make sure that:

• Cesium Viewer works.
• Works in 2D/CV.
• Works (or fails gracefully) in IE11.

[hpinkos]

@mramato I think this is ready. Did you have any other comments? Should we add a note to CHANGES.md?

[mramato]

Sorry, I meant to tag @lilleyse since I think he was the one that originally confirmed the issue was request/tile related and knows this area of the code. Sean?

[lilleyse]

I should have a chance to review later tonight.

[lilleyse]

Looks good to me. Nothing to add from what @mramato said. I opened a companion PR to fix this at the Heap/ManagedArray level in case other systems begin to use these files. #8889

@itsegg could you check if this branch fixes your issue in #4549 (comment)?

[itsegg]

@lilleyse
I think it's ok,
I have tested bn-dignitas:8843-fix-request-memory-retention,
My problem can be solved effectively.

[lilleyse]

Thanks @bn-dignitas!

[lilleyse]

Updated CHANGES.md in master: 2abec98

[mramato]

@lilleyse I have a regression introduced by this change, but it may be an unrelated bug that this is just covering up, not sure. Either way it's a RequestScheduler issue. I'm trying to create a simple reproduction now and then I'll write it up. Hopefully it's an easy fix (since you know this code way better than I do).