fix: TURBO_D=128 independent of QK_TURBO3, file turbo4 bugs PrismML-Eng#29

Codex post-commit review found:
1. TURBO_D was QK_TURBO3 (now 32) — broke turbo4 C array sizes
2. SET_ROWS kernel turbo3-specific but instantiated for turbo4
3. Tail block drop for non-128 head dims

Fixed PrismML-Eng#3 (TURBO_D). Mintplex-Labs#1 and Mintplex-Labs#2 don't affect turbo3+dk128 path.

Co-Authored-By: tturney@psyguard.ai
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: TheTom

ggml/src/ggml-turbo-quant.c

@@ -19,7 +19,7 @@
 
 #define TURBO_SEED_ROTATION 42
 #define TURBO_SEED_QJL      1042
-#define TURBO_D             QK_TURBO3  /* 128 — block size = head_dim */
+#define TURBO_D             128  /* rotation group size = head_dim (independent of block size) */
 #define TURBO_QJL_CONST     1.2533141373155003f  /* sqrt(pi/2) */
 
 /* Optimal centroids from paper (scaled by 1/sqrt(d)) */