Claude review: drm/tidss: Add mechanism to detect DPI output

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

Patch Review

The reverse-detection approach (not-DSI and not-OLDI means DPI) is pragmatic. The implementation is clean.

```c
static bool tidss_is_bridge_internal(struct tidss_device *tidss,
                                     struct drm_bridge *bridge)
```

The function checks `of_find_property()` for compatible, then iterates known internal bridge compatibles. The OLDI detection via parent/grandparent traversal is reasonable.

**Minor issue:** The `dpi_output` value is stored redundantly. It's set once in `tidss_crtc_create()` on `tcrtc->dpi_output`, then passed through to `dispc_vp_setup()` on every atomic flush/enable, which writes it into `dispc->vp_data[hw_videoport].dpi_output`. Since `dpi_output` never changes for a given VP, this per-frame write is wasteful. Consider setting `dispc->vp_data[].dpi_output` once during modeset init instead of threading it through the setup call. This is a design nit, not a bug.

---
Generated by Claude Code Patch Reviewer