Drop some dead code from FilterEffect

Source/platform/graphics/filters/FEDisplacementMap.cpp

Index: Source/platform/graphics/filters/FEDisplacementMap.cpp
diff --git a/Source/platform/graphics/filters/FEDisplacementMap.cpp b/Source/platform/graphics/filters/FEDisplacementMap.cpp
index 571fdc4eafafd2a56cd5e04d2bdd9264ca27f24a..f5f5e46c48337107947e7b753513d9b4209f632a 100644
--- a/Source/platform/graphics/filters/FEDisplacementMap.cpp
+++ b/Source/platform/graphics/filters/FEDisplacementMap.cpp
@@ -93,15 +93,6 @@ bool FEDisplacementMap::setScale(float scale)
     return true;
 }
 
-void FEDisplacementMap::setResultColorSpace(ColorSpace)
-{
-    // Spec: The 'color-interpolation-filters' property only applies to the 'in2' source image
-    // and does not apply to the 'in' source image. The 'in' source image must remain in its

[Stephen White, 2015/04/20 15:35:43]

This points out that there may be a bug in
FEDisplacementMap::createImageFilter(). Likely it should be passing
inputEffect(0)->operatingColorspace() to the first call to builder->builder(),
not operatingColorSpace().

If so, we're probably missing test coverage here.

[fs, 2015/04/20 15:59:29]

Yes, I noticed this discrepancy too... The commit that added this code [1], did
add tests too though, so I guess we need to check why that didn't notice it...

[1]
https://chromium.googlesource.com/chromium/blink/+/7d24fcc63dadc6a9e16159615e...

-    // current color space.
-    // The result is in that smae color space because it is a displacement of the 'in' image.
-    FilterEffect::setResultColorSpace(inputEffect(0)->resultColorSpace());
-}
-
 static SkDisplacementMapEffect::ChannelSelectorType toSkiaMode(ChannelSelectorType type)
 {
     switch (type) {