The C language has a basic flaw: it is impossible to return arrays from functions.
There are many workarounds for this; i'll describe three.
Replace by a pointer to an array
Return a pointer instead of an array itself. This leads to another problem in C: when a function returns a pointer to something, it should usually allocate the something dynamically. You should not forget to deallocate this later (when the array is not needed anymore).
typedef int (*pointer_to_array)[6][6];
pointer_to_array workaround1()
{
    pointer_to_array result = malloc(sizeof(*result));
    (*result)[0][0] = 0;
    (*result)[1][0] = 0;
    (*result)[2][0] = 0;
    (*result)[3][0] = 0;
    (*result)[4][0] = 0;
    (*result)[5][0] = 0;
    return result;
}
Replace by a pointer to int
A 2-D array appears just as a sequence of numbers in memory, so you can replace it by a pointer to first element. You clearly stated that you want to return an array, but your example code returns a pointer to int, so maybe you can change the rest of your code accordingly.
int *workaround2()
{
    int temp[6][6] = {{0}}; // initializes a temporary array to zeros
    int *result = malloc(sizeof(int) * 6 * 6); // allocates a one-dimensional array
    memcpy(result, temp, sizeof(int) * 6 * 6); // copies stuff
    return result; // cannot return an array but can return a pointer!
}
Wrap with a structure
It sounds silly, but functions can return structures even though they cannot return arrays! Even if the returned structure contains an array.
struct array_inside
{
    int array[6][6];
};
struct array_inside workaround3()
{
    struct array_inside result = {{{0}}};
    return result;
}