Given an m x n matrix, if an element is 0, set its entire row and column to 0. Do it in place and return the matrix.
Find every row and column that contains a zero first, then apply the zeroing in a second pass — mutating while you scan would zero out cells that should still be read as data.
Example
Input: matrix = [[1, 1, 1], [1, 0, 1], [1, 1, 1]]
Output: [[1, 0, 1], [0, 0, 0], [1, 0, 1]]