The estimated contribution to sea-level rise by thermal expansion of the oceans is 1 mm/year.
It's still not clear how this amount plus the measured contributions of Greenland (under 0.5 mm/year) and Antarctica (0.3 mm, possibly -0.2 mm) somehow add up to a predicted 6 mm/year (or even to a measured 3 mm/year trend).
1. Every time you burn a hydrocarbon, in addition to CO2 you also produce H2O. The ratio varies, but in general all of the hydrogen atoms will burn off before all of the carbon atoms do, since carbon-carbon bonds are fairly strong. (This can lead to things like soot.) Much of the water produced ultimately ends up in the oceans.
2. For decades now we've been pumping down aquifers at a fast clip, and that water ultimately ends up in the oceans, too. Natural aquifer replenishment rates may be fairly slow, if they ever replenish at all (the oldest, deepest aquifers may not) and the land-use changes we've made (paving and construction and such) slow that process down ever more.
3. Water from deep subterranean sources maybe be leaking into the oceans via undersea vents and such; this can also lead to bottom-up warming. Ongoing isostatic rebound from the loss of ice mass over the continents may be increasing these factors.
As to ocean warming itself, in general top-down warming only occurs within the uppermost layer of the ocean. (Oceans are dynamic things, of course, so some deeper flow and mixing does occur.) But that top-layer warming is also heavily mitigated by evaporation, where a great deal of the potential heat energy rises up with the water vapor and is eventually lost to space; meanwhile the water itself eventually returns. The clouds produced as the water vapor condenses and loses its heat also mitigate further warming.
If we're cooking up stories, how about this arbitrary "explanation": Phytoplankton are plentiful. They're reproducing and dying and sinking in such vast quantities to the ocean floor that this floor is steadily growing a new layer, at a rate of a few mm per year. Hence the apparent sea-level rise, relative to the continents. The accumulation of biogenic sediment is accelerating with the rise in atmospheric CO2 because CO2 directly feeds the plankton.
I consider the first type a form of education-induced brain damage, BTW. And while I'm fairly well-educated myself, I've never been the type who needed the "crutch" of a study in order to figure fairly obvious things out.
I don't remember the details, but I did once try to do the math on the whole CO2:H2O thing, since I hadn't seen it elsewhere. Estimated CO2 emission numbers are routinely published (to the extent that these can be trusted), and I started off with the simplifying assumption that CO2 and H2O are generated in equal amounts. This led to a small but real potential effect on sea level rise. (Although when I once discussed this in another forum some years back, someone there tried to insist that all of that water vapor just stays in the air - which I thought was rather bizarre!) But when I tried to make a more realistic assessment of the situation, I realized just how variable things could be (see examples below), and that I really had no good information to go on. Any numbers I came up with could vary widely based on my starting assumptions, so they weren't really worth very much in the end - kind of like the way so much "climate science" is done today! :)
CO2:H2O production ratios, if fully oxidized. If not fully oxidized then the CO2 numbers drop and soot is produced instead.
Methane: 1:2
Octane: 8:9
Benzene: 2:1
These are for very simple organic molecules. Once you get past a handful of carbon atoms the number of potential molecular configurations and corresponding ratios starts growing dramatically, and for something like coal the situation can grow extremely complex.
I know your phytoplankton comment was meant as sarcasm, but such things might actually have a real effect over time. (You should probably do the math!) The buildup of organic matter, minerals, sediments from erosion, and "rocks" and such might very well be noticeable. Remember that, if land-building forces were to basically come to a halt, then eventually all of the continents would pretty much just erode completely away, leaving not much more than an Earth covered in seawater to some average depth.
Estimates for biogenic sediment accumulation ranges up to 5 cm per thousand years -- two orders of magnitude smaller than current prediction of the rate of global sea-level rise.
A transition away from coal and towards natural gas (which consists primarily of fully-combustible methane at a ratio of 1:2) should lead to both an increase in H2O output and a decrease in CO2 output as compared to coal. (A recent local power plant transition from coal to natural gas claims a carbon emissions reduction of 60%.) Using your 2.5% number, water vapor is already about 60x more common in the atmosphere than CO2 is at 0.041%, and I've seen estimates that it is probably about 15% higher now than it was in the past. (Presumably any H2O output above that has ended up in the oceans.)
Given that as a greenhouse gas water vapor is at least 2x (IIRC) as powerful as CO2 and apparently also increasing at a steady clip, you'd think that folks would be paying more attention to this, but instead they just tend to ignore it. In fact they just like to ignore water vapor completely, except maybe when it leads to clouds and such.
Check page 291 for an overview of components and time lines. This is still the current scientific consensus AFAIK (no major updates).