[Test Improver] fix(tests): strip ANSI escape codes in test_policy_status._ascii_only

[danielmeppiel]

🤖 Test Improver — automated AI assistant

Problem

test_policy_status.py had 7 consistently failing tests. The _ascii_only() helper was returning False because Rich outputs ANSI CSI escape sequences (\x1b[3m italic, \x1b[1m bold, \x1b[0m reset) even inside Click's CliRunner, which does not set a pseudo-TTY. The ESC byte (\x1b, codepoint 0x1B) is categorised as Cc (control character) by unicodedata, so the existing guard correctly rejects it — but incorrectly flags Rich's styling codes as a violation.

Failures:

FAILED test_policy_status.py::TestStatusFoundOutcome::test_renders_found_outcome
FAILED test_policy_status.py::TestStatusAbsentOutcome::test_renders_absent_cleanly
FAILED test_policy_status.py::TestStatusCachedStaleOutcome::test_renders_stale_with_refresh_error
FAILED test_policy_status.py::TestStatusAsciiOnly::test_renderings_are_ascii_safe[found-...]
FAILED test_policy_status.py::TestStatusAsciiOnly::test_renderings_are_ascii_safe[absent-...]
FAILED test_policy_status.py::TestStatusAsciiOnly::test_renderings_are_ascii_safe[cached_stale-...]
FAILED test_policy_status.py::TestStatusAsciiOnly::test_renderings_are_ascii_safe[disabled-...]
```

## Approach

- Add `_strip_ansi()` helper using `re.sub(r"\x1b\[[0-9;]*[A-Za-z]", "", text)` — strips CSI escape sequences only, leaving actual text intact.
- Call `_strip_ansi()` at the top of `_ascii_only()` before iteration.
- The intent of `_ascii_only` is to catch non-ASCII *authored* strings (emojis, smart quotes) in CLI output — Rich's ANSI styling is not authored text and should be excluded from the check.

## Test Status

| Suite | Before | After |
|---|---|---|
| `tests/unit/commands/test_policy_status.py` | 7 failed, 33 passed | **40 passed** |
| Full unit suite | blocked at first failure | **6564 passed** |

```
uv run pytest tests/unit/commands/test_policy_status.py -q
# 40 passed in 0.34s

uv run pytest tests/unit tests/test_console.py -q
# 6564 passed, 1 warning, 26 subtests passed in 24.27s

Trade-offs

• Minimal change: one import (re), one compiled regex, one helper function.
• The regex \x1b\[[0-9;]*[A-Za-z] covers standard CSI sequences; OSC/DCS sequences are not stripped but don't appear in Rich table output.

Generated by Daily Test Improver · ● 1.9M · ◷

To install this agentic workflow, run

gh aw add githubnext/agentics/workflows/daily-test-improver.md@b87234850bf9664d198f28a02df0f937d0447295