feat(cli): Extend CLI with v1 authorization commands

[stuartcarnie]

Closes #19765

This commit add a series of new subcommands to the influx CLI tool. The specific commands are:

• influx v1 to host any InfluxDB v1 specific commands
• influx v1 auth for v1 authorization token management.
• influx v1 auth set-password to assign a password to a token for V1 API authorization

The v1 auth commands enable access to the private authorization APIs (see #19815 and #19837) used to manage authorization tokens for v1 compatibility APIs, including:

• /write
• /query

The final issue, #19768 will update the authorization handlers to use these new tokens and their assigned passwords.

• CHANGELOG.md updated with a link to the PR (not the Issue)
• Well-formatted commit messages
• Rebased/mergeable
• Tests pass

[stuartcarnie]

@gavincabbage the team-assigner gods picked you!

[stuartcarnie]

@gavincabbage I'll request a review when this is ready, as we need to make a few additional changes

[stuartcarnie]

@gavincabbage this PR is ready for review

[gavincabbage]

Looks good Stuart- just wanted to ask, would it be appropriate to add some tests for all the business logic in v1_authorization.go?

[stuartcarnie]

would it be appropriate to add some tests for all the business logic in v1_authorization.go?

I'm not really sure we have established a way to test CLI commands and I don't see precedent for testing any of the other ones. We do test the HTTP APIs that the CLI commands call, however.

[stuartcarnie]

Well, it looks like bucket has some tests:

influxdb/cmd/influx/bucket_test.go

Line 33 in 6bc4158

t.Run("create", func(t *testing.T) {

[stuartcarnie]

I'll take a closer look at the bucket tests and verify what they are testing – if they add value, I'll do the same.

[stuartcarnie]

@gavincabbage I'm going to add a follow up task to investigate refactoring these commands so they are testable.

The v1_authorization command was copied from authorization.go and is not testable, as the dependencies are not injected.

[gavincabbage]

Great, sounds good @stuartcarnie. I just made a similar call to skip testing something difficult to test until I refactor it to make it testable, so I know how that goes 😅