You don't have a product problem, you have a fear-of-launching problem. Stacks don't fail products, products fail because nobody knew about them. Pick the stack you have right now, ship to production with that, get 10 paying users, then decide if the stack matters. Spoiler: at that point you'll never want to migrate.
I've rebuilt the same SaaS in 3 stacks instead of launching. Call out the real problem, fear of launching, not tech. Then give me a no-migration plan: ship the current build to production as-is, get 10 paying users, and only revisit the stack after that. Give me the smallest concrete next step.