Terraform update

[zpatrick]

What does this pull request do? The main point of this PR is to update the terraform plugin. I've updated all of the code to work with the update client / job model. I've also slightly changed the unit test patterns for this package. While testing this, I encountered quite a few bugs/missing features needed to properly test these changes. With that, I've also fixed the following bugs / added the missing functionality:

• Adds tags for jobs
• Adds a custom error handler to the api, so all uncaught errors are still returned as a ServerError. It also maps our error codes to meaningful http status codes.
• Fixes cli text printer tests and implementation for tasks
• Adds retry logic to service create, since that would sometimes fail with the "failed to assume role and validate ports" error, due to the loadbalancer still coming up.
• Adds an optional scale param to service create
• Adds the missing job command to the cli
• Adds shorthand flag names for output and timeout, e.g. l0 -o json ...

How should this be tested? Unit test all of the package where I have made changes. To test the terraform provider, I usually just create a main.tf with every available resource and data source. I add/update/remove each of the resources' options until i've exhausted all possible configurations. You also need to test that it gracefully handles when things no longer exist - e.g. it was created in terraform but deleted by the CLI.

Checklist

• [x] Unit tests
• [x] Documentation (if applicable)

[diemonster]

Why are there so many vendored code changes in this PR? Are they necessary for terraform or were they auto-updated as part of something else?

[zpatrick]

Why are there so many vendored code changes in this PR? Are they necessary for terraform or were they auto-updated as part of something else?

I believe this was because I ran go format on the entire repo

[sesh-kebab]

@zpatrick would you please be able to update develop.md with development workflow information for the terraform provider also?

[diemonster]

@sesh-kebab @zpatrick to seshi's point, I think what we need detailed here eventually is "How should this be tested?". This is still very WIP though so I don't think it's fully fair to have that info yet.

[diemonster]

Is there any use for the Layer0 admin API in the terraform plugin?

[jparsons04]

Can you explain the test pattern where you create a resource in terraform and then delete the resource via the cli. Are you saying that if you were to terraform plan after the deletion that terraform would gracefully handle that situation? I didn't see what the specific testing pattern was supposed to be from terraform's perspective for gracefully handling that.