Claude review: drm/bridge: analogix_dp: Extract error pointer from correct variable

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

Patch Review

**Bug Analysis**: Correct and real bug. The code does:
```c
dp->plat_data->next_bridge = devm_drm_panel_bridge_add(dp->dev,
                                                       dp->plat_data->panel);
if (IS_ERR(dp->plat_data->next_bridge)) {
    ret = PTR_ERR(bridge);  // BUG: wrong variable
    goto err_unregister_aux;
}
```

`bridge` is a separate local variable (the `struct drm_bridge *` for `dp->bridge`) that is a valid pointer at this point. Calling `PTR_ERR()` on a valid pointer returns garbage — typically a large positive number that gets truncated to an `int`, resulting in a misleading or zero return code. The caller would not see a proper `-ENODEV`/`-ENOMEM`/etc.

**Fix**: Correct. Changes `PTR_ERR(bridge)` to `PTR_ERR(dp->plat_data->next_bridge)`, matching the variable tested by `IS_ERR()`.

```c
ret = PTR_ERR(dp->plat_data->next_bridge);  // FIXED: matches IS_ERR check
```

**Commit message**: Clear and well-structured. Includes the Smatch warning, a proper `Fixes:` tag referencing the commit that introduced the bug (`1b86a69b61df4`), and a signed-off-by.

**Minor notes**:
- The `Fixes:` tag hash `1b86a69b61df4` uses 13 hex chars — the standard convention is 12 chars. This is cosmetic and most tooling handles it fine, but `1b86a69b61df` would be more conventional.
- As noted above, this code has been removed in drm-next. The maintainers should consider whether this patch should target stable branches instead.

**Rating**: The fix is correct, minimal, and addresses a genuine bug. Recommended for merge (targeting the appropriate branch where this code still exists).

---
Generated by Claude Code Patch Reviewer