lavfi/ebur128: advance pointer to samples

Correct the advancement of a pointer to skip over LFE samples as
intended in the loudness computation.

This suggested patch is 3 of 3 for Ticket #2144 "libavfilter ebur128
loudness inaccuracy, irregular time interval, LFE interference".
This suggested patch supersedes the contribution of January 15, 2013
in the thread “[PATCH] lavfi: EBU R.128 irreg time, ch weights, skip
LFE” which combined multiple fixes in one batch.

libavfilter/f_ebur128.c

@@ -478,6 +478,8 @@ static int filter_frame(AVFilterLink *inlink, AVFilterBufferRef *insamples)
         for (ch = 0; ch < nb_channels; ch++) {
             double bin;
 
+            ebur128->x[ch * 3] = *samples++; // set X[i]
+
             if (!ebur128->ch_weighting[ch])
                 continue;
 
@@ -491,8 +493,6 @@ static int filter_frame(AVFilterLink *inlink, AVFilterBufferRef *insamples)
                                       - dst[1]*name##_A1 - dst[2]*name##_A2;    \
 } while (0)
 
-            ebur128->x[ch * 3] = *samples++; // set X[i]
-
             // TODO: merge both filters in one?
             FILTER(y, x, PRE);  // apply pre-filter
             ebur128->x[ch * 3 + 2] = ebur128->x[ch * 3 + 1];