Add LOGGER_ASSERT for checking fatal error conditions.

[iphydf]

These are not compiled out under NDEBUG and should be provably correct.

---

This change is 

[hugbubby]

Reviewable status: 0 of 1 approvals obtained (waiting on @iphydf)

---

toxcore/logger.c, line 107 at r1 (raw file):

        log = &logger_stderr;
#else
        fprintf(stderr, "NULL logger not permitted\n");

Since this is a direct fprintf statement, punctuation would be appropriate, no?

---

toxcore/tox.c, line 1135 at r1 (raw file):

set_message

I get that the case is here for documentation purposes but it's still superfluous

[hugbubby]

Reviewable status: 1 change requests, 0 of 1 approvals obtained (waiting on @iphydf)

---

toxcore/tox.c, line 1135 at r1 (raw file):

Previously, hugbubby wrote…

set_message

I get that the case is here for documentation purposes but it's still superfluous

(meant to point to case -5: and not set_message)

[iphydf]

Reviewable status: 1 change requests, 0 of 1 approvals obtained (waiting on @hugbubby)

---

toxcore/logger.h, line 103 at r1 (raw file):

Previously, hugbubby wrote…

Capitalize the A here

Done.

---

toxcore/logger.c, line 107 at r1 (raw file):

Previously, hugbubby wrote…

Since this is a direct fprintf statement, punctuation would be appropriate, no?

Done.

---

toxcore/tox.c, line 1135 at r1 (raw file):

Previously, hugbubby wrote…

(meant to point to case -5: and not set_message)

Done.

[hugbubby]

Reviewed 3 of 3 files at r2.
Reviewable status: 1 change requests, 0 of 1 approvals obtained

[hugbubby]

Wouldn't know what Travis is complaining about, but I approved it because it LGTM