Refuse a clipped highlight as a neutral

Sampling the overcast sky on a Canon 6D frame set tint to -100 and
temperature to -15 for a patch the canvas showed as pure white. A clipped
photosite is sensor white, not a colour: every channel stopped counting,
so what the tap hands back is the as-shot multipliers themselves, which
are strongly magenta, and the solver dutifully drove green to its stop.
The display shader already fades such a pixel to a neutral of the same
brightness before any operation runs, so the picker was balancing against
something the photographer could not see.

The probe now refuses a sample with any channel at or above the onset the
shader fades from, the way the solver already refuses black. The threshold
is one constant, CLIP_ONSET, formatted into the shader and read by the
probe, so the two cannot drift apart.

ui/dr-ui/src/develop.rs

@@ -4502,6 +4502,15 @@ impl DevelopSession {
         if pixel[3] < 0.5 {
             return None;
         }
+        // Nor in a clipped one. A blown sky reads as sensor white, and sensor
+        // white with the as-shot balance on is strongly magenta — a solve over
+        // it drives tint to its stop for a pixel that, on the canvas, the
+        // shader has already desaturated to neutral. The same threshold the
+        // shader fades from, so what is refused here is what it would have
+        // hidden there.
+        if pixel[..3].iter().any(|c| *c >= dr_pipeline::CLIP_ONSET) {
+            return None;
+        }
 
         // The tap is the sensor's numbers with the profile filled neutral;
         // the operation multiplies them *after* the camera's own balance, so
@@ -6745,6 +6754,52 @@ mod tests {
         );
     }
 
+    /// TRACES: FR-DEV-3
+    /// A blown highlight is refused, the way black is.
+    ///
+    /// Sensor white is not a colour: every channel stopped counting, so the
+    /// ratio between them is the as-shot multipliers and nothing about the
+    /// scene. Sampling the overcast sky on a Canon 6D frame drove tint to
+    /// -100 and temperature to -15 for a patch the canvas showed as pure
+    /// white, which is the picker being wrong rather than the point being a
+    /// poor choice. Refused, nothing moves and no step is taken.
+    #[test]
+    fn sampling_a_blown_highlight_moves_nothing() {
+        let Some(ctx) = headless() else { return };
+        let size = 16u32;
+        let raw = RawImage {
+            width: size,
+            height: size,
+            data: vec![65535; (size * size * 3) as usize],
+            cfa_pattern: dr_decode::CfaPattern::Rggb,
+            black_level: [0; 4],
+            white_level: 65535,
+            wb_coeffs: [1.9, 1.0, 1.7, 0.0],
+            color_matrix: None,
+            base_curve: dr_decode::BaseCurve::IDENTITY,
+            samples_per_pixel: 3,
+            profile: None,
+            make: String::new(),
+            model: String::new(),
+            crop: dr_decode::CropRect {
+                x: 0,
+                y: 0,
+                width: size,
+                height: size,
+            },
+        };
+        let mut session =
+            DevelopSession::open(&ctx, &raw, dr_types::Orientation::NORMAL).expect("session");
+        let steps = session.history_rows().len();
+
+        assert!(
+            !session.sample_neutral(0.5, 0.5),
+            "a clipped photosite has no balance in it"
+        );
+        assert!(session.is_neutral(), "and so nothing was corrected");
+        assert_eq!(session.history_rows().len(), steps);
+    }
+
     /// TRACES: FR-DEV-7 | FR-DEV-5
     /// A held comparison hands the edit straight back.
     ///