flaky test: test_operator_gpu.test_conv

[eric-haibin-lin]

test_operator_gpu.test_conv ... [00:29:53] c:\jenkins_slave\workspace\build-gpu\src\operator\nn\cudnn\./cudnn_algoreg-inl.h:107: Running performance tests to find the best convolution algorithm, this can take a while... (setting env variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
[00:29:53] C:/jenkins_slave/workspace/build-gpu/src/operator/nn/convolution.cu:148: This convolution is not supported by cudnn, MXNET convolution is applied.
[00:29:53] C:/jenkins_slave/workspace/build-gpu/src/operator/nn/convolution.cu:227: This convolution is not supported by cudnn, MXNET convolution is applied.
[00:29:53] C:/jenkins_slave/workspace/build-gpu/src/operator/nn/convolution.cu:148: This convolution is not supported by cudnn, MXNET convolution is applied.
[00:29:53] C:/jenkins_slave/workspace/build-gpu/src/operator/nn/convolution.cu:227: This convolution is not supported by cudnn, MXNET convolution is applied.
[INFO] Setting test np/mx/python random seeds, use MXNET_TEST_SEED=1207962136 to reproduce.
[INFO] Setting test np/mx/python random seeds, use MXNET_TEST_SEED=2083930564 to reproduce.
ERROR

http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/master/1043/pipeline

http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/PR-11360/17/pipeline

[vrakesh]

@eric-haibin-lin Thank you for reporting the Flaky test, this is added to the to do test improvements