chore(ci): unblock CI by gating bash tests on tree-sitter; sort review prompt glob

[elfensky]

Summary

CI has been red on main since 2026-04-06, and on every PR opened since, because of 5 pre-existing test failures unrelated to any feature work. Because they live on main, every PR inherits them and can't go green without bundling a workaround (see #614, #616). This standalone, test-only PR fixes the root cause so main gets a green baseline.

Two distinct root causes:

1. Bash tests assume tree-sitter is installed

desloppify/tests/lang/common/test_bash_unused_imports.py calls detect_unused_imports, which parses via tree-sitter. tree-sitter is an optional dependency ([project.optional-dependencies]), so the tests-core job (no extras) doesn't have it — there detect_unused_imports returns [] and the tests, asserting specific findings, fail. In tests-full (pip install -e ".[full]") tree-sitter is present and the tests pass.

Fix: gate the file with the exact skipif pattern its sibling tree-sitter tests already use (test_treesitter.py, test_treesitter_complexity_and_integration.py):

pytestmark = pytest.mark.skipif(
    not is_available(), reason="tree-sitter-language-pack not installed"
)

This is a consistency fix — the bash file was the only tree-sitter-dependent test file missing the guard. Coverage is preserved: with tree-sitter (tests-full) the tests run and assert real findings; without it (tests-core) they skip cleanly, because the feature genuinely cannot run there.

2. Review tests rely on filesystem-dependent glob order

test_review_commands.py (and its integration counterpart) use list(runs_dir.glob("*/prompts/batch-*.md")) unsorted. macOS happens to return batch-1.md first; Linux returns batch-2.md first. The assertion "Previously flagged issues" in prompt_text only holds for batch-1.md, so it fails on Linux.

Fix: sorted(...) the glob. Deterministic across OSes, removes a latent flake that only passed on macOS by luck, and the assertion still verifies exactly what it intends (batch-1's content).

Why this is safe

• Test-only. No product/runtime behaviour changes.
• No tests made meaningless. Fix 1 preserves full bash coverage in tests-full and only skips where the optional feature is absent; the graceful-degradation path is itself covered by test_treesitter.py::test_is_available_false_when_uninstalled. Fix 2 strengthens (de-flakes) its test.

Verification

Reproduced both CI environments locally:

• tree-sitter present (mirrors tests-full): the 3 bash tests run and pass, both review tests pass.
• tree-sitter absent (mirrors tests-core): the bash tests skip with reason tree-sitter-language-pack not installed; review fix is OS-independent.

🤖 Generated with Claude Code