fix(pitch_shift): re-anchor fadePos_ on freeze so gNew is continuous at the freeze frame; guard left cast; add step-detector + loop-never-freezes regression

src/vst/pitch_shift.cpp

@@ -105,15 +105,27 @@ void PitchShifter::freezeTail() {
     // An in-flight crossfade was sized for a RETREATING writer (outgoing tap drains at
     // ratio-1 per frame); frozen, the outgoing tap closes at the full ratio. Cap the live
     // fade so it completes before tap B reaches the parked writer and reads lapped (oldest-
-    // window) content mid-fade. `+1` keeps fadeLen_ > fadePos_, so t stays < 1 in process().
+    // window) content mid-fade. fadePos_ is re-anchored to the same fractional t so gNew is
+    // continuous at the freeze frame (no gain step); see the re-anchor block below.
     if (fading_) {
+        // Preserve t = fadePos_/fadeLen_ across the shortening so gNew is continuous at the
+        // freeze frame (no gain step). Compute tOld BEFORE overwriting fadeLen_, then
+        // re-anchor fadePos_ to the same fractional position in the new (shorter) fade.
+        const double tOld =
+            static_cast<double>(fadePos_) / static_cast<double>(fadeLen_);
         double dB = static_cast<double>(writePos_) - posB_;
         const double len = static_cast<double>(ringLen_);
         while (dB < 0.0) dB += len;
         while (dB >= len) dB -= len;
-        const double left = (dB - 2.0) / ratio_;
+        // Clamp in double before the int64 cast (matches splice() pattern; guards against UB
+        // when dB/ratio_ is very large, e.g. near-unity ratio at a high sample rate).
+        double left = (dB - 2.0) / ratio_;
+        if (left > static_cast<double>(fadeFrames_)) left = static_cast<double>(fadeFrames_);
         const std::int64_t leftFrames = left > 1.0 ? static_cast<std::int64_t>(left) : 1;
-        fadeLen_ = std::min(fadeLen_, fadePos_ + leftFrames);
+        const std::int64_t newFadeLen = std::min(fadeLen_, fadePos_ + leftFrames);
+        // Re-anchor: tOld < 1 because we are mid-fade, so newFadePos < newFadeLen (still fading).
+        fadePos_ = static_cast<std::int64_t>(tOld * static_cast<double>(newFadeLen));
+        fadeLen_ = newFadeLen;
     }
 }