Follow-up fix for parsing non-utf8 chars (part 2)

[Gumix]

This patch is similar to commit 52d3e4f ("Follow-up fix for parsing non-utf8 chars"),
but fixes the --verbose output, e.g.:

$ ./test-run.py --verbose some_test

[...]
[001] Worker "001_xxx-luatest" received the following error; stopping...
[001] Traceback (most recent call last):
[001]   File "./tarantool/test-run/lib/worker.py", line 357, in run_task
[001]     short_status, duration = self.suite.run_test(
[001]                              ^^^^^^^^^^^^^^^^^^^^
[001]   File "./tarantool/test-run/lib/test_suite.py", line 271, in run_test
[001]     short_status = test.run(server)
[001]                    ^^^^^^^^^^^^^^^^
[001]   File "./tarantool/test-run/lib/test.py", line 232, in run
[001]     color_stdout(f.read(), schema='log')
[001]                  ^^^^^^^^
[001]   File "<frozen codecs>", line 322, in decode
[001] UnicodeDecodeError: 'utf-8' codec can't decode byte 0xa3 in position 115111: invalid start byte
[001]
[001] Exception: 'utf-8' codec can't decode byte 0xa3 in position 115111: invalid start byte

Follow-up #442

[coveralls]

coverage: 62.656% (-0.04%) from 62.696%
when pulling 395d619 on iverbin/fix-non-utf8-chars-parsing
into f23b535 on master.