use well known order of lerp arguments

amd.hh

@@ -32,7 +32,7 @@ public:
 	value_type operator()(complex_type input)
 	{
 		value_type amp = abs(input);
-		avg = lerp(amp < avg ? dec : att, avg, amp);
+		avg = lerp(avg, amp, amp < avg ? dec : att);
 		avg = min(max(avg, lo), hi);
 		amp /= avg;
 		return (amp * (value_type(1)+idx) - value_type(1)) / idx;

ema.hh

@@ -35,7 +35,7 @@ public:
 	}
 	TYPE operator()(TYPE input)
 	{
-		return prev = lerp(alpha, prev, input);
+		return prev = lerp(prev, input, alpha);
 	}
 };
 
@@ -83,7 +83,7 @@ public:
 	TYPE operator()(TYPE input)
 	{
 		for (int i = 0; i < ORDER; ++i)
-			prev[i] = input = lerp(alpha, prev[i], input);
+			prev[i] = input = lerp(prev[i], input, alpha);
 		return input;
 	}
 };

utils.hh

@@ -16,8 +16,8 @@ int signum(TYPE v)
 	return (v > TYPE(0)) - (v < TYPE(0));
 }
 
-template <typename X, typename AB>
-AB lerp(X x, AB a, AB b)
+template <typename AB, typename X>
+AB lerp(AB a, AB b, X x)
 {
 	return (X(1) - x) * a + x * b;
 }