Add --json flag for resource-class cmds

[jdelnano]

Checklist

=========

• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have checked for similar issues and haven't found anything relevant.
• This is not a security issue (which should be reported here: https://circleci.com/security/)
• I have read Contribution Guidelines.

Internal Checklist

Not applicable to me, a non-CircleCI employee

• I am requesting a review from my own team as well as the owning team
• I have a plan in place for the monitoring of the changes that I am making (this can include new monitors, logs to be aware of, etc...)

Changes

=======

Implementation

I implemented my changes in the following way:

• I added a local --json flag to the resource_class.go file where it was appropriate to, for the create and list command actions.
• In each command, I check whether or not the --json flag was set on the command line, and if so, we Marshal the appropriate object and set it in the Writer interface offered by the cobra cmd object.
• I introduced the JSON formatting logic such that the original CLI behavior is unchanged. If you submit commands like:
  • circleci runner resource-class list <>
  • circleci runner resource-class create <>

they will behave exactly as they did. Only if you include a --json flag will you see new behavior, returned to stdout.

Tests

I made sure to add tests in resource_class_test.go where appropriate as well as update the [create-expected-usage.txt](https://github.com/CircleCI-Public/circleci-cli/compare/develop...jdelnano:circleci-cli:add-json-output-flag-for-resource-class-cmds?expand=1#diff-694751a2b7d7a982b3a13b6bcb00d307f3e25dc7014c17520056d3d3c1dbd8ca) file.

Rationale

=========

Background

In my position at Okta, a CircleCI customer, I have a few automation tasks that require the use of the circleci runner resource-class list and circleci runner resource-class create commands, where I specifically need to grab the output of these commands and do subsequent actions. As a result, my motivation for adding the --json flag support for the circleci runner resource-class command is to return more easily parsable output, instead of the default ASCII table format.

Manual Testing

I executed the following commands manually (alongside the automated testing), to ensure:

• They work as expected, adding in the new JSON formatting.
• The old CLI behavior is maintained as before these changes.

make build

➜  circleci-cli git:(add-json-output-flag-for-resource-class-cmds) ./build/darwin/arm64/circleci runner resource-class list <circleci-org>
➜  circleci-cli git:(add-json-output-flag-for-resource-class-cmds) ./build/darwin/arm64/circleci runner resource-class list <circleci-org> --json
➜  circleci-cli git:(add-json-output-flag-for-resource-class-cmds) ./build/darwin/arm64/circleci runner resource-class create <circleci-org>/<resource-class-name> "test"
➜  circleci-cli git:(add-json-output-flag-for-resource-class-cmds) ./build/darwin/arm64/circleci runner resource-class create <circleci-org>/<resource-class-name> "test" --json
➜  circleci-cli git:(add-json-output-flag-for-resource-class-cmds) ./build/darwin/arm64/circleci runner resource-class create <circleci-org>/<resource-class-name> "test" --generate-token
➜  circleci-cli git:(add-json-output-flag-for-resource-class-cmds) ./build/darwin/arm64/circleci runner resource-class create <circleci-org>/<resource-class-name> "test" --generate-token --json

Screenshots

============

Before

Image or [gif](https://giphy.com/apps/giphycapture)

After

Image or gif where change can be clearly seen

Here are some helpful tips you can follow when submitting a pull request:

1. Fork the repository and create your branch from main.
2. Run make build in the repository root.
3. If you've fixed a bug or added code that should be tested, add tests!
4. Ensure the test suite passes (make test).
5. The --debug flag is often helpful for debugging HTTP client requests and responses.
6. Format your code with gofmt.
7. Make sure your code lints (make lint). Note: This requires Docker to run inside a local job.