fix(rebase): adapt fmt=4 to dev's quant.h refactor + colibri.c rename

Dev refactored glm.c → colibri.c and extracted the matmul/quant kernels
into quant.h (matmul, matmul_q, matmul_i4, matmul_i4_grouped, matmul_i2,
quant_scratch, dot_i4i8, matmul_q_idot, matmul_i4_idot, etc. all live
there now). The original #298 commit re-added all of these inline; on
rebase they became duplicate definitions.

Resolution:
- Removed the ~700-line duplicate block (everything dev moved to quant.h)
- Kept ONLY the unique fmt=4 contribution: matmul_i4_grouped_pair (the
  fused gate+up kernel that reads x once instead of twice, ~33% decode
  speedup) + the fmt=4 branch in expert_gate_up that dispatches to it.
  Dev's expert_gate_up only fused fmt==2; this adds the fmt==4 case.
- Forward-declared matmul_i4_grouped_pair before expert_gate_up.
- Fixed quant_matmul call site in the ragged attention path (backend_cuda.cu)
  to pass gs/ng — the kernel signature gained those args in the attention
  scales fix, but dev's new ragged path called it with the old signature.

Build-verified: colibri.exe (CPU + COLI_CUDA) and coli_cuda.dll both
compile clean on the rebased branch.

c/backend_cuda.cu

@@ -1122,7 +1122,7 @@ extern "C" int coli_cuda_attention_project_ragged(ColiCudaTensor *w,ColiCudaTens
     attention_absorb_ragged_kernel<<<dim3(H,S),256,shared,dc->stream>>>(dc->ac,dc->aq,ddl,ddr,
         dn,w->weights,w->scales,w->fmt,S,H,Q,R,V,K,T,scale);
     quant_matmul<<<dim3(proj->O,S),256,0,dc->stream>>>(dc->y,dc->ac,proj->weights,
-        proj->scales,proj->fmt,S,proj->I,proj->O,row_bytes(proj->fmt,proj->I));
+        proj->scales,proj->fmt,S,proj->I,proj->O,row_bytes(proj->fmt,proj->I),proj->gs,proj->ng);
     return cuda_ok(cudaGetLastError(),"ragged attention launch")&&
            cuda_ok(cudaMemcpyAsync(out,dc->y,ob,cudaMemcpyDeviceToHost,dc->stream),"ragged output download")&&
            cuda_ok(cudaStreamSynchronize(dc->stream),"ragged attention synchronize");