Add TF CIFAR test to continuous testing set

[laurenyu]

Description of changes:
Though this test has historically been relatively flaky, there was a recent change to the training steps that helped stabilize it, so it should be fine to include this test with our continuous testing. This test is important because it's currently our only one that tests distributed training and the only one to run on GPU instances.

Merge Checklist

Put an x in the boxes that apply. You can also fill these out after creating the PR. If you're unsure about any of them, don't hesitate to ask. We're here to help! This is simply a reminder of what we are going to look for before merging your pull request.

• I have read the CONTRIBUTING doc
• I have added tests that prove my fix is effective or that my feature works (if appropriate)
• I have updated the changelog with a description of my changes (if appropriate)
• I have updated any necessary documentation (if appropriate)

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

[codecov-io]

Codecov Report

Merging #170 into master will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #170   +/-   ##
=======================================
  Coverage   90.63%   90.63%           
=======================================
  Files          37       37           
  Lines        2478     2478           
=======================================
  Hits         2246     2246           
  Misses        232      232

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update de954b0...af5e24c. Read the comment docs.

[nadiaya]

It's the test that takes longest time to run, right?
sigh