Add ARM Grace Blackwell (NVIDIA DGX Spark) support (#922)

This commit enables IQK quantization operations on ARM-based systems,
specifically tested on NVIDIA DGX Spark with GB10 Grace Blackwell.

Changes:
- Enable IQK_IMPLEMENT macro for ARM NEON operations
- Add arm_neon.h header include for ARM SIMD intrinsics
- Fix compilation errors related to missing NEON types and functions

Build requirements for ARM:
  cmake .. -DGGML_CUDA=ON \
           -DCMAKE_CXX_FLAGS="-march=armv8.2-a+dotprod+fp16" \
           -DCMAKE_C_FLAGS="-march=armv8.2-a+dotprod+fp16"

Tested on:
- Platform: NVIDIA DGX Spark (aarch64)
- CPU: GB10 Grace Blackwell Superchip
- Memory: 128GB unified memory

Fixes build errors:
- 'float32x4_t' does not name a type
- 'vld1q_f32' was not declared in this scope
- 'v_expf' was not declared in this scope
- Missing FP16 NEON intrinsics

ggml/src/iqk/iqk_cpu_ops.cpp

@@ -4,6 +4,8 @@
 // SPDX-License-Identifier: MIT
 //
 
+#define IQK_IMPLEMENT
+
 #include "iqk_cpu_ops.h"
 #include "iqk_utils.h"
 #include "ggml.h"
@@ -14,6 +16,10 @@
 #include <cmath>
 #include <cstring>
 
+#ifdef __ARM_NEON
+#include <arm_neon.h>
+#endif
+
 namespace {
 // Playing around with group scores: use sum of probabilities in the group
 inline float group_score(int n_per_group, const float * data) {