fix(rtl): add use_dsp="no" attributes, fix duplicate wire declaration

Add (* use_dsp = "no" *) to all modules containing multiplication
operators to force LUT-based multiplication instead of DSP inference:
- barrett_mul.v (3 multipliers x 7 instances = 21 ops)
- comp_decomp_sync.v, sample_ntt_sync*.v, mlkem_top.v

Also fix duplicate wire declaration of ct_bytes_rt in mlkem_top.v.

sync_rtl/comp_decomp/comp_decomp_sync.v

@@ -10,6 +10,7 @@
 
 `include "sync_rtl/common/defines.vh"
 
+(* use_dsp = "no" *)
 module comp_decomp_sync (
     input         clk,
     input         rst_n,

sync_rtl/ntt/barrett_mul.v

@@ -6,6 +6,7 @@
 // Pure combinational, single-cycle latency.
 // All multiplication widths explicitly controlled to avoid Verilog truncation.
 
+(* use_dsp = "no" *)
 module barrett_mul (
     input  [11:0] a,
     input  [11:0] b,

sync_rtl/sample_ntt/sample_ntt_sync.v

@@ -31,6 +31,7 @@
 `include "sync_rtl/common/defines.vh"
 
 /* verilator lint_off UNUSEDPARAM */
+(* use_dsp = "no" *)
 module sample_ntt_sync #(parameter K = 4) (
 /* verilator lint_on UNUSEDPARAM */
     input              clk,

sync_rtl/sample_ntt/sample_ntt_sync_shared.v

@@ -44,6 +44,7 @@
 `include "sync_rtl/common/defines.vh"
 
 /* verilator lint_off UNUSEDPARAM */
+(* use_dsp = "no" *)
 module sample_ntt_sync_shared #(parameter K = 4) (
 /* verilator lint_on UNUSEDPARAM */
     input              clk,

sync_rtl/top/mlkem_top.v

@@ -22,6 +22,7 @@
 
 `include "sync_rtl/common/defines.vh"
 
+(* use_dsp = "no" *)
 module mlkem_top #(
     parameter KMAX = 4               // storage sizing (worst case = ML-KEM-1024)
 ) (
@@ -133,7 +134,6 @@ module mlkem_top #(
     wire [11:0] c1_bytes_rt = 12'd32 * {7'b0, du_rt} * {9'b0, k_r};  // 640/960/1408
     wire [11:0] c2_bytes_rt = 12'd32 * {7'b0, dv_rt};                 // 128/128/160
     wire [11:0] ct_bytes_rt = c1_bytes_rt + c2_bytes_rt;             // 768/1088/1568
-    wire [11:0] ct_bytes_rt = c1_bytes_rt + c2_bytes_rt;             // 768/1088/1568
     // Shared secret out: Encaps drives ss_r = K directly. Decaps implicit reject:
     // ss = (c' == c) ? K' : K-bar. K' is in ss_r, K-bar in kbar_r; dec_reject is
     // the latched c' != c flag from D7 (0 for KeyGen/Encaps, so ss_r passes through).