There are some data type inconsistencies in the tbl impl functions.

[y-vectorfield]

When I built the source codes of BitNet.cpp using gcc, some data type inconsistencies error occured.
However, this error did not occur when I used clang. I investigated the root cause of this problem.
I found that there are some data type inconsistence in the tbl impl functions.
The followin codes contain some data type inconsisitencies problem.

BitNet/utils/codegen_tl1.py

Lines 204 to 207 in caf17ec

	vec_c[{6}] += vec_v_left_{0}.val[0];\n\
	vec_c[{6}] += vec_v_right_{0}.val[0];\n\
	vec_c[{7}] += vec_v_left_{0}.val[1];\n\
	vec_c[{7}] += vec_v_right_{0}.val[1];\n\

BitNet/utils/codegen_tl1.py

Line 235 in caf17ec

const int8x16_t vec_zero = vdupq_n_s16(0x0000);\n\

BitNet/utils/codegen_tl1.py

Line 252 in caf17ec

vec_c[i] = vandq_s16(vec_c[i], vec_zero);\n\

We should add some data type reinterpreting cast processes in the codes.