Rule idea: Prefer `monkeypatch.setenv` over direct `os.environ` mutation in tests

### Summary

Directly mutating `os.environ` in tests can leak state across test cases and cause order-dependent failures. Use pytest’s `monkeypatch.setenv`, which automatically restores environment variables after the test completes, ensuring proper isolation and reproducibility.

### Example

```python
import os
import pytest

def foo() -> str:
return os.environ.get("ENV_VAR", "default_value")

# Bad
def test_bad():
os.environ["ENV_VAR"] = "value"
assert foo() == "value"

# Good
def test_good(monkeypatch: pytest.MonkeyPatch):
monkeypatch.setenv("ENV_VAR", "value")
assert foo() == "value"
```

The issue names pytest's monkeypatch.setenv and direct os.environ mutation but no repository files or tests. Start by locating Ruff's existing rule definitions and test cases for Python environment-variable usage, then determine how the rule should identify the mutation and verify that compliant monkeypatch.setenv usage is accepted.

Written by the indexing model from the issue text.