refactor: Change workspace list to use view.View instead of cli.Ui when rendering human output to the terminal

[SarahFrench]

Background

Part of #37439

This PR changes just the workspace list command to use the newer view.View abstraction for rendering output. Please see the issue above for more context.

The workspace list command only supports human-readable output, before this PR and after this PR. However the changes in this PR make adding support for machine-readable output reasonably easy. I'm planning to do that in future, separately.

This PR

This PR introduces structs in the views package (WorkspaceHuman, etc) to enable rendering diagnostics and text notices to the terminal. Diagnostics can be received from backends/state stores when retrieving the list of workspaces via their Workspaces methods or could be raised in the command itself (#38094). The text notices include the list output of the command itself:

* default
  dev
  stage
  prod

Those structs are then used in the workspace select command. E.g. instead of c.Ui.Error to log errors, the code uses the generic view.Diagnostics method to render diagnostics (both warning or error level).

Things for the reviewer to note

There is a change in behaviour when rendering warning diagnostics.

When using cli.Ui, warnings are logged to stderr:

1. When the tool is used E2E: https://github.com/hashicorp/cli/blob/4383e52914f5677576e148a6d7e1f399c0e91a7c/ui_common.go#L76-L78
2. When we mock the Ui in tests: https://github.com/hashicorp/cli/blob/4383e52914f5677576e148a6d7e1f399c0e91a7c/ui_mock.go#L75-L80

when using view.View's Diagnostics method, warnings are logged to stdout:

terraform/internal/command/views/view.go

Lines 120 to 124 in ac3e32b

	if diag.Severity() == tfdiags.Error {
	v.streams.Eprint(msg)
	} else {
	v.streams.Print(msg)
	}

I'm still trying to work out if we consider this a breaking change or not, but I'm leaning towards not:

• Workspace commands are not used in the context of HCPTF, so there's no concern of breaking how CLI output is consumed there (Slack me for context on this).
• We technically don't have any guarantees for any human output

The only issue I've found so far is that the parsing of human output in terraform-exec relies on stdout containing only the list of workspaces. I've got a proposed change to fix this in: hashicorp/terraform-exec#562
However the best route forward would be to add machine-readable output support to the workspace commands and update terraform-exec to use that instead.

Target Release

N/A

Rollback Plan

• If a change needs to be reverted, we will roll out an update to the code within 7 days.

Changes to Security Controls

Are there any changes to security controls (access controls, encryption, logging) in this pull request? If so, explain.

CHANGELOG entry

• This change is user-facing and I added a changelog entry.
• This change is not user-facing.

[github-actions]

Changelog Warning

Currently this PR would target a v1.15 release. Please add a changelog entry for in the .changes/v1.15 folder, or discuss which release you'd like to target with your reviewer. If you believe this change does not need a changelog entry, please add the 'no-changelog-needed' label.

[SarahFrench]

To be removed during review, but leaving here for now so it's super obvious where we can see the impact of this change so far

[SarahFrench]

Closing this - I have another branch with these types of changes but allow continued use of cli.UI for human output to prevent breaking changes.

[github-actions]

I'm going to lock this pull request because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active contributions.
If you have found a problem that seems related to this change, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.