Claude review: drm/bridge: Add ->detect_ctx hook and drm_bridge_detect_ctx()

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

Patch Review

This patch adds the atomic variant of `drm_bridge_funcs.detect()` and a new `drm_bridge_detect_ctx()` helper. The implementation is clean and follows the established patterns from the connector-level `detect_ctx`.

**Looks good.** A few observations:

The fallback path when `detect_ctx` is NULL but `detect` exists is correct:
```c
return bridge->funcs->detect(bridge, connector);
```

The self-managed locking path (ctx == NULL) properly handles EDEADLK:
```c
drm_modeset_acquire_init(&br_ctx, 0);
retry:
    ret = drm_modeset_lock(&connector->dev->mode_config.connection_mutex, &br_ctx);
    if (!ret)
        ret = bridge->funcs->detect_ctx(bridge, connector, &br_ctx);
    if (ret == -EDEADLK) {
        drm_modeset_backoff(&br_ctx);
        goto retry;
    }
```

The conversion of negative errors to `connector_status_unknown` is appropriate:
```c
if (ret < 0)
    ret = connector_status_unknown;
```

**Minor doc nit:** The doc says "If @ctx is set, it might also return -EDEADLK" — but when ctx is passed through directly to `detect_ctx`, the callback could return *any* negative error, not just -EDEADLK. The caller would see all of them. This is handled correctly in practice (patch 2's `drm_bridge_connector_detect_ctx` checks `ret < 0`), but the doc could be more precise.

---
Generated by Claude Code Patch Reviewer