Retry network operations in resources scripts

[heaths]

Relates to Azure/azure-sdk-tools#326 and other issues.

[heaths]

/azp run java - storage - tests

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[heaths]

@danieljurek the failure seems to be unrelated, but I'd hate for it to go unchecked:

https://dev.azure.com/azure-sdk/public/_build/results?buildId=236312&view=logs&j=2714c728-d141-5cd1-e4fc-49513190fa65&t=4da905a1-2f38-556a-c21b-c51eb108a69d&l=33061

[heaths]

/azp run java - storage - tests

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[heaths]

Live tests aren't reporting back status like in other repos for some reason, but the only failure(s) seem to be unrelated and most often not even connecting to Azure, though are numerous: https://dev.azure.com/azure-sdk/internal/_build/results?buildId=236460&view=logs&j=351a4efb-edfe-5bb4-cda1-1ace5fc54b68&t=59f8b7da-4461-55d1-d907-5626116ef0aa&l=2716. Other platform/JRE combinations passed. The "Windows - Java 8" job also appears like it will time out, but that's also not uncommon from past runs, it seems.

[heaths]

Had a Java dev look over and agreed none of this would cause the failures we're seeing in the live tests.