avcodec/pixlet: Fix shift exponent 4294967268 is too large for 32-bit type 'int'

Fixes: 1336/clusterfuzz-testcase-minimized-4761381930795008

Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/targets/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
(cherry picked from commit 1f5b6c7e1e)
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

libavcodec/pixlet.c

@@ -154,7 +154,7 @@ static int read_low_coeffs(AVCodecContext *avctx, int16_t *dst, int size, int wi
         state = 120 * (escape + flag) + state - (120 * state >> 8);
         flag = 0;
 
-        if (state * 4 > 0xFF || i >= size)
+        if (state * 4ULL > 0xFF || i >= size)
             continue;
 
         nbits = ((state + 8) >> 5) + (state ? ff_clz(state) : 32) - 24;
@@ -260,7 +260,7 @@ static int read_high_coeffs(AVCodecContext *avctx, uint8_t *src, int16_t *dst, i
 
         flag = 0;
 
-        if (state * 4 > 0xFF || i >= size)
+        if (state * 4ULL > 0xFF || i >= size)
             continue;
 
         pfx = ((state + 8) >> 5) + (state ? ff_clz(state): 32) - 24;