fresnel-fourier iter7 Phase 7 fix-forward: data links connect pads not entities directly

Empirical Phase 7 verification revealed the algorithm bug: data links
in MEDIA_IOC_G_TOPOLOGY connect PAD IDs, not entity IDs directly.
My iter7 Phase 6 commit compared link source_id/sink_id against
the proc entity_id, never matched → io_entity_ids stayed empty →
interface lookup never fired → returns -1 → falls back to legacy
hardcoded path.

Topology dump on fresnel /dev/media0 (rkvdec) confirmed:
- Entity 3 (rkvdec-proc) has function=0x4008 (DECODER) ✓
- Data link src=16777218 sink=16777220 — these are PAD ids
  (0x01000002, 0x01000004), NOT entity 3.
- Interface link src=50331660 (interface) sink=1 (entity) — for
  interface links source/sink ARE entity IDs.

Fix: resolve pads → entities via the topo.pads[] array.
1. Collect pads belonging to proc entity (via pads[].entity_id).
2. For each data link touching those pads, the OTHER pad's
   entity_id is an IO neighbor.
3. Find interface link to those IO entities (unchanged from prev).

Also allocate topo.pads[] in the 2-call ioctl pattern.

Signed-off-by: claude-noether <claude-noether@reauktion.de>

tests/run_perf_binding_cell.sh

@@ -171,17 +171,19 @@ run_consumer() {
 	sleep 1
 
 	# Parse pidstat by header: locate the %CPU column index from the
-	# column-name row, then apply it to data rows. Robust across
-	# sysstat 12.x point releases (where column positions shift).
+	# column-name row (where any field equals "%CPU"), then apply it
+	# to data rows. Robust across sysstat 12.x point releases.
+	# pidstat default output has no '#' header marker — the header is
+	# the first row containing "%CPU" as a field.
 	awk '
-		# Header row: find which field is %CPU.
-		$1 == "#" {
-			for (i = 1; i <= NF; i++) if ($i == "%CPU") col = i
-			next
+		# Header row: any line where some field equals "%CPU".
+		!col {
+			for (i = 1; i <= NF; i++) if ($i == "%CPU") { col = i; next }
 		}
-		# Data row: skip the average summary at end + blank lines.
-		col && NF >= col && $1 ~ /^[0-9]/ {
-			if ($col ~ /^[0-9]+(\.[0-9]+)?$/) print $col
+		# Data row: lines whose value at $col is numeric. Skip the
+		# trailing "Average" summary by requiring $col to parse cleanly.
+		col && NF >= col && $col ~ /^[0-9]+(\.[0-9]+)?$/ {
+			print $col
 		}
 	' "$logdir/pidstat.log" >"$logdir/cpu_pct.log" || true