refactor: rename the node from opencv-plot to dora-plot

The repository is dora-plot, but the package, the entrypoint, the manifest
name and the window title were all opencv-plot. Consumers wire the node by
`path:`, so the split name was visible in every dataflow.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018WLVZCFFcCH1EpTXGfS9aD

dora_plot/bbox.py

@@ -0,0 +1,37 @@
+"""Parsing of the `bbox` input: an Arrow struct carrying boxes and labels."""
+
+import numpy as np
+
+from dora_plot.plot import Bboxes
+
+
+def parse(value, bbox_format: str) -> Bboxes:
+    """Read a `bbox` payload. `bbox_format` is the `format` metadata parameter."""
+    fields = value[0]
+    corners = _flat(fields, "bbox").reshape(-1, 4)
+
+    if bbox_format == "xyxy":
+        pass
+    elif bbox_format == "xywh":
+        # xywh is centre-based: x and y are the box centre, not a corner.
+        x, y, w, h = corners.T
+        corners = np.stack([x - w / 2, y - h / 2, x + w / 2, y + h / 2], axis=1)
+    else:
+        raise RuntimeError(f"unsupported bbox format: {bbox_format}")
+
+    return Bboxes(
+        corners=corners,
+        conf=_flat(fields, "conf"),
+        labels=_flat(fields, "labels", numeric=False),
+    )
+
+
+def _flat(fields, name: str, numeric: bool = True) -> np.ndarray:
+    """Read a struct field as a flat numpy array.
+
+    Senders wrap each field in a single-element list -- that is what
+    `pa.array([{...}])` produces from numpy arrays -- so unwrap it when present.
+    """
+    column = fields[name]
+    column = getattr(column, "values", column)
+    return column.to_numpy(zero_copy_only=False) if not numeric else column.to_numpy()