Forest for the trees, again.
Paladin 5 has 4 first level spells and 2 second level spells. Warlock 2, paladin 3 has 5 first level spells (2 of which recharge on a SR), and zero second level spells. To say that the problem does not exist AT ALL, when the problem was being behind in spell progression every other level, is missing something. Here you're always behind in highest level spellcasting by 2 full levels of your class spellcasting. For full casters (Sorlock), that is a full spell level of spells at all times. For a half-caster (Hexadin), that is going to be half the impact.
In either instance: Electing to multiclass has a cost, the cost is easy to see, and it discourages min-maxing approaches to combine classes by making it slightly less efficienct to spell cast as compensation for getting a broader range of spells, etc...
As for the question of why did they require errata after all of those hours of playtesting? Most of the errata is clarification and fine tuning. Light hammer rather than throwing hammer is not exactly a major deal. Expecting perfection on the fine tuning is a bit ridiculous. The errata is almost entirely a bit more editing that took place after the hundreds of thousands of hours of playtesting had grown to billions of hours of actual use.
Or do you see a swatch of significant rebalancing that I am missing?