Use the crypto random functions instead of rand().

[iphydf]

Presumably the uses of rand() were fine because they were not used in
security-sensitive places, but having to think about whether a crappy RNG
is acceptable in each situation requires effort that could better be
spent elsewhere.

Also, this means that once we have a custom deterministic RNG for
testing, that RNG is used everywhere, so all the code is deterministic.

It also allowed us to delete a system-specific function that wasn't used
anywhere except in a call to srand().

---

This change is 

[sudden6]

Reviewed 8 of 8 files at r1.
Reviewable status: 0 of 1 LGTMs obtained (waiting on @zugz)

[robinlinden]

Reviewed 8 of 8 files at r1.
Reviewable status: 0 of 1 LGTMs obtained (waiting on @zugz)

[zugz]

Reviewable status: 0 of 1 LGTMs obtained (waiting on @zugz)

---

toxcore/onion_client.c, line 270 at r2 (raw file):

    }

    uint32_t num_nodes = (onion_c->path_nodes_index < MAX_PATH_NODES) ? onion_c->path_nodes_index : MAX_PATH_NODES;

Might as well make it const?

---

toxcore/onion_client.c, line 1624 at r2 (raw file):

                    for (j = 0; j < n; ++j) {
                        uint32_t num = random_u32() % num_nodes;

const?

---

toxcore/onion_client.c, line 1744 at r2 (raw file):

            if (num_nodes != 0) {
                for (i = 0; i < (MAX_ONION_CLIENTS_ANNOUNCE / 2); ++i) {
                    uint32_t num = random_u32() % num_nodes;

const?

[iphydf]

Reviewable status: 0 of 1 LGTMs obtained (waiting on @zugz)

---

toxcore/onion_client.c, line 270 at r2 (raw file):

Previously, zugz (zugz) wrote…

Might as well make it const?

Done.

---

toxcore/onion_client.c, line 1624 at r2 (raw file):

Previously, zugz (zugz) wrote…

const?

Done.

---

toxcore/onion_client.c, line 1744 at r2 (raw file):

Previously, zugz (zugz) wrote…

const?

Done.