Merge pull request #27596 from pratham-mcw:core-kmeans-loop-unroll

core: ARM64 loop unrolling in kmeans to improve Weighted Filter performance #27596

### Pull Request Readiness Checklist

See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request

- [x] I agree to contribute to the project under Apache 2 License.
- [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [x] The PR is proposed to the proper branch

- This PR improves the performance of the Weighted Filter function from the ximgproc module on Windows on ARM64.
- The optimization is achieved by unrolling two performance-critical loops in the generateCentersPP function in modules/core/src/kmeans.cpp, which is internally used by the Weighted Filter function.
- The unrolling is enabled only for ARM64 builds using #if defined(_M_ARM64) guards to preserve compatibility and maintain performance on other architectures.

**Performance Improvements:** 
- Improves execution time for Weighted Filter performance tests on ARM64 without affecting other platforms.
<img width="772" height="558" alt="image" src="https://github.com/user-attachments/assets/ae28c0af-97d3-460b-ad5a-207d3fc6936f" />

modules/core/src/kmeans.cpp

@@ -131,7 +131,15 @@ static void generateCentersPP(const Mat& data, Mat& _out_centers,
                           KMeansPPDistanceComputer(tdist2, data, dist, ci),
                           (double)divUp((size_t)(dims * N), CV_KMEANS_PARALLEL_GRANULARITY));
             double s = 0;
-            for (int i = 0; i < N; i++)
+            int i = 0;
+        #if CV_ENABLE_UNROLLED
+            for (; i + 7 < N; i += 8)
+            {
+                s += tdist2[i + 0] + tdist2[i + 1] + tdist2[i + 2] + tdist2[i + 3] +
+                     tdist2[i + 4] + tdist2[i + 5] + tdist2[i + 6] + tdist2[i + 7];
+            }
+        #endif
+            for (; i < N; i++)
             {
                 s += tdist2[i];
             }