Claude review: drm/ttm/tests: fix lru_count ASSERT

[Claude Code Review Bot <claude-review@example.com>]

Patch Review

**The bug:** After pool initialization via `__list_lru_init -> init_one_lru`, the LRU count for each node should be zero. The original assertion had a spurious `!` negation:

```c
KUNIT_ASSERT_FALSE(test,
                   !list_lru_count(&pt.pages));
```

`KUNIT_ASSERT_FALSE(test, !list_lru_count(...))` means "assert that `!count` is false", which is equivalent to "assert count is **non-zero**". That's the opposite of what the test intended — it should be verifying the LRU is empty after initialization.

**The fix:** Removing the `!` gives:

```c
KUNIT_ASSERT_FALSE(test,
                   list_lru_count(&pt.pages));
```

This correctly asserts that `list_lru_count(&pt.pages)` is zero (falsy), matching the expected post-init state.

I confirmed that the kernel tree at `drivers/gpu/drm/ttm/tests/ttm_device_test.c:178-179` already has the corrected version (without the `!`), consistent with this fix having been applied on drm-next.

**Minor style observation:** Using `KUNIT_ASSERT_EQ(test, list_lru_count(&pt.pages), 0)` would make the intent unambiguous and avoid future confusion with negation, but the existing form is fine and consistent with current test style.

**Reviewed-by worthy:** Yes. The fix is correct, minimal, and well-described. The `Fixes:` tag is appropriate (references `444e2a19d7fd` which ported ttm/pool to list_lru and introduced this test).

---
Generated by Claude Code Patch Reviewer