The final parsec problem describes the difficulty of two blackholes merging with one another when their distances reach ~3.2 light years apart. Does this also then apply to stellar collisions, perhaps with another distance threshold? If not, why?
Potentially related?: What is the difference between the terms collision and merger? How are they used differently in Astronomy?
The "final parsec problem" describes the difficulty in getting two supermassive black holes close enough together that their merger timescale due to gravitational wave emission becomes shorter than the age of the universe.
Normal stars do not merge due to gravitational wave emission - their masses are too low and their separations too large. The gravitational wave merger timescales would be much longer than the age of the universe even when their photospheres are in contact.
There is something akin to the final parsec problem for merging stellar black holes, neutron stars and white dwarfs. These are compact enough to have mergers driven by gravitational wave emission but they too must get together closer than about $10^{10}$ m (for black holes) or $10^9$ m (for neutron stars and white dwarfs) before the gravitational waver merger timescales become shorter than 10 billion years. However, this is very much a solved problem in the sense that their are lots of channels whereby this can happen. Common envelope evolution in isolated binary systems can bring binary components together (e.g., Kruckow et al. 2018). Multiple dynamical interactions within star clusters (Sedda 20121) or as part of ternary or other multiple systems can also do the trick (Vynethya & Hamers 2022).
