feat(issue): add lc issue status command

[bougyman]

Summary

• Adds lc issue status ISSUE_ID to change an issue's workflow state
• --status/-s sets the state by name (case-insensitive exact match, then unique prefix); prompts interactively when omitted
• --comment/-m adds a comment alongside the status change
• "s" subcommand alias mirrors other single-letter issue aliases (c, d, l, u)
• Resolves ISSUE_ID via the existing expand_issue_id/1 convention
• Errors clearly on unknown or ambiguous status names (exit 22 via {:smells_bad, …})
• JSON output emits the updated issue struct only; text mode adds a confirmation line

Implementation

• New :set_status Ash action on LinearCli.Linear.Issue with a dedicated LinearCli.Linear.Issue.Update.SetStatus manual update module (mirrors the pattern of :close, :assign, :attach_to_project)
• set_issue_status/2 domain code interface in LinearCli.Linear
• status subcommand entry in the Optimus spec with ISSUE_ID positional arg and --status/--comment options
• Commands.issue_status/1 orchestrates the full flow: expand id → fetch issue → fetch team states → resolve target state → optional comment → transition → display

Test plan

• Exact name match (case-insensitive via --status done)
• Short -s flag
• Unique prefix match (e.g. --status in → "In Progress")
• Unknown status name → exit 22 + clear message
• Ambiguous prefix → exit 22 + clear message
• --comment/-m adds comment before transitioning
• Interactive selection (no --status): prompts sorted by position
• --output json emits parseable JSON only
• "s" alias routes correctly

Closes https://linear.app/cryptokairos/issue/CRY-47/add-lc-issue-status-command

🤖 Generated with Claude Code

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[bougyman]

Addressed the review comment: replaced the nested case in resolve_target_state/2 with the pipeline-based approach you suggested. Now splits into three named helpers — use_prefix_matches_if_empty/3 (exact-match fallback) and resolve_state_matches/3 (three-clause pattern match on list shape) — eliminating the duplicated ambiguous-match branch and the cyclomatic complexity. All 25 issue_commands tests still pass.

[bougyman]

Fixed the CI failure: ran mix format on the two files that were flagged (commands.ex and issue_commands_test.exs). Long map literals and one long if were broken onto multiple lines per the formatter's 98-column rule. All 25 issue_commands tests still pass. Commit: d274a68.

[bougyman]

Code Review (Run 4)

Branch: CRY-47-add-issue-status-command (3 commits: b8a7b7e, 11d27c7, d274a68)
Reviewer: Independent code review agent — no prior context on this issue.

Summary

The implementation adds lc issue status ISSUE_ID backed by a new :set_status Ash action. Code is clean, idiomatic, and follows established project patterns. All 248 non-environment-specific tests pass. Formatting and compilation are clean.

Issues Found

None critical. None major.

Minor:

1. Dead stub_responses call in test (app/test/linear_cli/cli/issue_commands_test.exs:559-561)
   The first test ("--status sets the workflow state by exact name") calls stub_responses/1 and then immediately overwrites it with a manual Req.Test.stub/2 two lines later. The stub_responses call is dead code — it has no effect. All other status tests correctly use only the manual stub. Suggest removing lines 559-561.

2. @doc says "New command." (app/lib/linear_cli/cli/commands.ex:479)
   The phrase "New command." is temporal — it will stop being "new" and become stale. Consider removing or replacing with just the behavior description.

3. No test for -m short form — The -s short form for --status has its own test, but -m for --comment does not. Optimus handles the mapping, so this is a framework guarantee, but adding a one-liner assertion would be consistent.

What Looks Good

• Pipeline decomposition of resolve_target_state/2 into use_prefix_matches_if_empty/3 + resolve_state_matches/3 is clean idiomatic Elixir — multi-clause functions replacing nested case statements.
• Ash action structure (SetStatus manual update module) follows the exact same pattern as Close, Assign, and AttachToProject. Consistent and composable.
• Test coverage is thorough: exact match, prefix match, ambiguous match, unknown status, interactive selection, comment flow, JSON output, and alias routing — all covered with correct assertions.
• Error semantics use {:error, {:smells_bad, msg}} consistently, mapping to exit 22 via the existing handle_error/3 path.
• Subcommand alias "s" => "status" doesn't conflict with any existing alias.

Pre-existing Observation (not introduced by this PR)

The {:ok, [issue]} <- Linear.issues(...) pattern in the with chain (line 492) will fall through as {:ok, []} if the issue ID doesn't resolve, causing a CaseClauseError in run/3. This is the same pattern used in gimme_da_issue!/2 (issue_helpers.ex:442) — a pre-existing gap, not a regression.

Overall Assessment

Approve. The three minor items are cleanup nits, not blockers. The feature is correct, well-tested, and follows project conventions.

[bougyman]

Rework (Run 5)

Addressed the two minor cleanup items from code review:

• app/lib/linear_cli/cli/commands.ex — Removed "New command." from @doc opener (temporal phrase that goes stale immediately)
• app/test/linear_cli/cli/issue_commands_test.exs — Removed dead stub_responses([...]) call in the first status test (was unconditionally overwritten by the Req.Test.stub two lines below it)

Commit: 0431f97 (style(issue-status): remove review-noted clutter)
All 25 issue_commands tests pass.