Eager CAPTURE format probe in RequestInit

Chromium's vaapi_video_decoder may call vaCreateContext with surfaces=NULL,
surfaces_count=0 and then create surfaces afterwards via vaCreateSurfaces2.
In that order driver_data->video_format is still NULL when CreateContext
runs and our early `video_format == NULL` guard returns OPERATION_FAILED.
Confirmed via temporary request_log() — Brave hits exactly that path on
ohm.

Move the probe out of RequestCreateSurfaces into a new
video_format_probe() helper in video.c, and call it eagerly from
RequestInit. RequestCreateSurfaces still re-probes if init came up NULL,
which preserves the original lazy behaviour for any caller that needs it.

Also small clean-up of surface.c since the probe block moved out: drop
the now-dead `bool found` local.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

src/request.c

@@ -39,6 +39,7 @@
 #include "request.h"
 #include "utils.h"
 #include "v4l2.h"
+#include "video.h"
 
 #include <assert.h>
 #include <stdio.h>
@@ -179,6 +180,19 @@ VAStatus VA_DRIVER_INIT_FUNC(VADriverContextP context)
 	driver_data->video_fd = video_fd;
 	driver_data->media_fd = media_fd;
 
+	/* Probe the CAPTURE pixel format eagerly so vaCreateContext doesn't
+	 * race ahead of vaCreateSurfaces. Fine if this returns NULL — we
+	 * still let the driver init succeed and let surface creation try
+	 * again later (preserves the original lazy behaviour for any caller
+	 * that needs it). */
+	driver_data->video_format = video_format_probe(video_fd);
+	if (driver_data->video_format != NULL)
+		request_log("Init: detected CAPTURE format %s (mplane=%d)\n",
+			    driver_data->video_format->description,
+			    driver_data->video_format->v4l2_mplane);
+	else
+		request_log("Init: no CAPTURE format detected at probe time\n");
+
 	status = VA_STATUS_SUCCESS;
 	goto complete;