[GPU] fix small error in fully_connected bf_tiled kernel (#32643)

### Details:
- found when trying to run quantized to u8 mobilenetv2-7 from
https://github.com/microsoft/onnxruntime-inference-examples/tree/main/quantization/image_classification/cpu

### Tickets:

Author: michal-miotk

src/plugins/intel_gpu/src/kernel_selector/cl_kernels/fully_connected_gpu_bf_tiled.cl

@@ -1024,7 +1024,7 @@ inline void FUNC(fc_bf_tiled_kernel_dyn_quan)(
         unroll_for(uint fi = 0; fi < TILE_OFM; ++fi) {
             #if DECOMPRESSION_ZP_TERM
                 #if DECOMPRESSION_ZP_SCALAR
-                    wei_zp[fi] = (TO_ACCUMULATOR_TYPE)(DECOMPRESSION_ZP_VALUE);
+                    wei_zp[fi] = TO_ACCUMULATOR_TYPE(DECOMPRESSION_ZP_VALUE);
                 #elif DECOMPRESSION_ZP_GROUPS_NUM == 1
                     wei_zp[fi] = TO_ACCUMULATOR_TYPE(d_zps[fi % DECOMPRESSION_ZP_LENGTH]);
                 #endif