Water ligands form dative covalent bonds with the central transition metal ion. That would mean a molecule is formed, just like sulfur dioxide or any other.
The transition metal ion does not form dipole ion attraction with water as does sodium chloride, but forms a covalent bond. If so, why is the resulting mixture a solution, and not a solid?
The transition metal ion does not form dipole ion attraction with water as does sodium chloride
Actually it does. What we refer to as the complex is only the innermost hydration shell. This comples is still positively charged so the water molecules around this complex will still tend to orient themselves oxygen-inwards to build up dipole-dipole interactions (and hydrogen bonds) with the proper complex. Likewise, you should not consider dissolved sodium ions as in sodium chloride any different. They, too, have a hydrate shell which can be described as a coordination complex. The key difference is that the hydrate shell is very labile and no precise geometry can be given. There is little energy cost associated with water molecules diffusing away or attaching back. But other than that, the two are the same.
Whether or not the complex ion (e.g. $\ce{[Fe(H2O)6]^2+}$) is in solution or in a solid depends on what you did with the compound. A lot of hydrates are actually cationic aquacomplexes; for example, solid $\ce{CuSO4.5H2O}$ can be described as $\ce{[Cu2(H2O)8(\mu{-}SO4)2] . 2H2O}$ instead, showing that we have complexes in the solid state, too.
