avcodec/g729postfilter: Limit shift in long term filter

Fixes: shift exponent 34 is too large for 32-bit type 'int'
Fixes: 57389/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_ACELP_KELVIN_fuzzer-6229522659016704

Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpegSigned-off-by: Michael Niedermayer <michael@niedermayer.cc>

libavcodec/g729postfilter.c

@@ -353,7 +353,7 @@ static int16_t long_term_filter(AudioDSPContext *adsp, int pitch_delay_int,
         if (tmp > 0)
             L_temp0 >>= tmp;
         else
-            L_temp1 >>= -tmp;
+            L_temp1 >>= FFMIN(-tmp, 31);
 
         /* Check if longer filter increases the values of R'(k). */
         if (L_temp1 > L_temp0) {