I see no reason to believe that we will have different numbers of Maritime, Cultured and Mercantile. Like you say, Mogadishu has been seen and would be likely to be either Maritime or Mercantile, so I guess we'll have more coming that we haven't seen yet for a total of 14-14-14-8-8, although 13-13-13-8-8 would also be a possibility (13/8 is closer to the old ratio of 10/6 than 14/8).
5 yet to be seen city states seems like a lot at this point, but always possible. If it's 13-13-13-8-8 it's only two though. Maybe they might have gone for an odd number?
As for the ratio. It was ~14.3% Militaristic and Religious in Gods & Kings, whilst for 13 of Mercantile, Maritime and Cultured it would be ~14.5% and 13.8% for 14.
Another way of looking at it would be Major types to Minor types ratio. That is, the fraction that is Militaristic and Religious (Minor Types). Here is a quick table:
Vanilla: 28.6%
Gods & Kings: 28.6%
12 Majors: 30.8% <- Already ruled out
13 Majors: 29.1%
14 Majors: 27.6%
So yes, having it 13-13-13-8-8 would be the closest to the Gods & Kings and Vanilla ratio in that case. In fact, this pretty much answers the old question about why they changed the ratio between Vanilla and Gods & Kings though. About as close as they could get in fact. Other ratios that could get close are:
10 Majors, 6 Minors: 28.6% < - Staying the same
15 Majors, 9 Minors: 28.6%
13 Majors, 8 Minors: 29.1%
11 Majors, 7 Minors: 28.0%
15-15-15-9-9 would have worked perfectly if in the aim was to keep the ratio the same, but 13-13-13-8-8 gets pretty close. Not sure they would have wanted to add 21 new city states though. Could still be the case though, that would be 9 unknown new city states remaining if it were the case, but statistically I don't see that happening as we really should have seen more new city states if that were the case. Of the 42 returning city states (replacements included), we've only not seen 5 of them, which would be 11.9%. We'd expect to see a similar number for the new city states as well, and as there are so far 11 of them (Mogadishu included) then we'd expect one or so more.
Another thought occurs though. Whilst 13-13-13-8-8 doesn't exactly conserve the ratio, 14-13-13-8-8 does, so that is another possibility. Anyhow, if this is the case, we'd expect:
15-15-15-9-9: 9 new city states <- Match for the old ratio
14-13-13-8-8: 3 new city states <- Match for the old ratio
13-13-13-8-8: 2 new city states
13-12-12-8-8: No new city states
With what is discussed above in mind, I'd have a guess between 14-13-13-8-8 and 13-13-13-8-8, I'd actually think that 14-13-13-8-8 could be quite likely in fact.