I am only an interested layman in cosmology and GR. In recent years, we have learned that black holes indeed exist in the universe and that almost all galaxies contain massive black holes in their centers. Black holes, in essence only devour other (stellar) matter and they also swallow each other as has been demonstrated by the gravitational wave discoveries. Is it possible that black holes eventually will "eat" most or even all normal and dark matter of the universe? And could this, and a merging of black holes, lead to a situation where all the universe ends up being a few or even only one black hole?
I think black holes have been misrepresented in popular culture. They aren't giant cosmic vacuum cleaners devouring everything in their path, they are gravitational objects which exist and have their own stable orbits. Things do fall into them, but they aren't sucking everything towards them at some extreme rate.
Ultimately most cosmologists believe the universe will keep expanding until everything tends toward absolute 0 temperature and there is no energy in a sense, the black holes that are there will exist until pretty much the end of time, but even those will probably slowly radiate away due to Hawking radiation until there is literally nothing. Black holes generally are too far away from each other to feel the gravitational influence of even their closest black hole neighbor so the likelihood of them all consuming each other and merging is extremely narrow, especially as with cosmic expansion they are getting further apart at all times.
No. There is a misconception that a lot of laypeople seem to be picking up recently from popularizations, which is that all the matter in the universe will first be sucked into black holes, and then recycled into photons in the distant future through Hawking radiation, so that the only thing left will be photons. This is wrong.
Black holes are not super-powerful cosmic vacuum cleaners that can suck in everything there is. When a black hole forms through gravitational collapse, such as the collapse of a dying star, its mass stays the same, and that means that at some fixed distance such as a billion kilometers, its gravity is no stronger than it used to be. A black hole is in fact a very small object, typically about the size of a city. This makes it a small target, and the space between the stars is vast, so a huge amount of luck is required if it is to swallow another astronomical object through a chance collision.
So if we look at the cosmic cereal box a gajillion years from now, what will be on the list of ingredients? It will contain stable, massive subatomic particles such as electrons.[Baez 2004] (The exact roster of particles depends on unknown particle physics such as dark matter and proton decay.) Any photons with ordinary energies, such as those originating from the present-day cosmic microwave background, will play a negligible role as their energy density becomes diluted faster than the mass-energy density of material particles. (This process has already been going on for billions of years. It's is the reason why the cosmic microwave background, which was dominant in the early universe, is now a negligible part of the universe's mass-energy.) However, the universe will contain photons with extremely low energies, which an observer (if one could be present and could detect them) would describe as Hawking radiation from the cosmological event horizon.[Hu 2011] These photons will have a temperature on the order of 10^-30 K, meaning a typical energy of 10^-34 eV and a wavelength of 10^28 m.
Adams and Laughlin, "A Dying Universe: The Long Term Fate and Evolution of Astrophysical Objects", Rev. Mod. Phys. 69 (1997) 337, http://arxiv.org/abs/astro-ph/9701131
Baez, "The End of the Universe," 2004, http://math.ucr.edu/home/baez/end.html
Penrose, Causality, quantum theory and cosmology. In On Space and Time, ed. Shahn Majid, Cambridge University Press, Cambridge, 2008, pp. 141-195. (ISBN 978-0-521-88926-1)
Hu, "Hawking radiation from the cosmological horizon in a FRW universe," Phys.Lett. B701 (2011) 269-274, http://arxiv.org/abs/1007.4044