Implement a program that copies an \(N \times N\) matrix of 32-bit floating point numbers from input array \(A\) to output array \(B\) on the GPU. The program should perform a direct element-wise copy so that \(B_{i,j} = A_{i,j}\) for all valid indices.
solve function signature must remain unchangedB
Input: A = [[1.0, 2.0],
[3.0, 4.0]]
Output: B = [[1.0, 2.0],
[3.0, 4.0]]
Input: A = [[5.5, 6.6, 7.7],
[8.8, 9.9, 10.1],
[11.2, 12.3, 13.4]]
Output: B = [[5.5, 6.6, 7.7],
[8.8, 9.9, 10.1],
[11.2, 12.3, 13.4]]
N ≤ 4096N = 4,096