Explore behavior of flakey R test UNET from test_img_seg.R

[DickJC123]

Description

Eager to get CI healthy. I'm no R expert, so just taking some stabs in the dark.

Checklist

Essentials

• PR's title starts with a category (e.g. [BUGFIX], [MODEL], [TUTORIAL], [FEATURE], [DOC], etc)
• Changes are complete (i.e. I finished coding on this PR)
• All changes have test coverage
• Code is well-documented

Changes

• Feature1, tests, (and when applicable, API doc)
• Feature2, tests, (and when applicable, API doc)

Comments

• If this change is a backward incompatible change, why must this change be made.
• Interesting edge cases to note here

[mxnet-bot]

Hey @DickJC123 , Thanks for submitting the PR
All tests are already queued to run once. If tests fail, you can trigger one or more tests again with the following commands:

• To trigger all jobs: @mxnet-bot run ci [all]
• To trigger specific jobs: @mxnet-bot run ci [job1, job2]

---

CI supported jobs: [clang, centos-cpu, unix-gpu, sanity, windows-cpu, centos-gpu, windows-gpu, unix-cpu, edge, miscellaneous, website]

---

Note:
Only following 3 categories can trigger CI :PR Author, MXNet Committer, Jenkins Admin.
All CI tests must pass before the PR can be merged.

[DickJC123]

@thirdwing Reaching out to you as an "R Ninja" on issue #19169.

This PR makes a (theoretically) non-functional change to the failing test UNET of test_img_seg.R. The modified test has passed CI 7 times with no error (not sure if this is enough times to draw a conclusion). Can you think of an explanation why this change might fix the problem? Could there be multiple threads or race conditions present?

[szha]

cc @hetong007. I'm merging this in to stabilize the CI first.