You have identified the difference between practical and theoretical optimization. Theoretical optimization identifies only the end product in the presence of a "neutral but benign GM." Practical optimization is worried about the paths and the playability at all levels.
It is easy to state this goal, of course. In practice, this means rebuilding a character four or five (or seventeen) times to account for the discrepancies introduced during the creation process. It also means finding a way to experience the class at various choke points to be aware of the optimization level at each choke point.
Looking at the Theurge you gave in your example, it is immediately obvious that the opportunity cost of a 2 level multiclass is nominally intolerable for the "Thou shalt never lose caster levels" Tier 1 classes. Therefore, we can look for solutions which minimize this opportunity cost. Ur-priest, the various early-entry methods, race selection to maintain effective caster level, etc.
None of these tricks are difficult nor even non-obvious. However, this problem illustrates my Constrained Optimization paper nicely: a good character is about well stated requirements. It is a common failing of the theoretical optimizer that they do not sufficiently articulate their requirements such that the character is playable or fun.
Practically, there are many ways to mitigate "low-level" weakness in a character. They all start with a coherent level by level build of the character, the level goals, and the intended play style. By creating quantifiable requirements, it is possible to anticipate "low-level" weaknesses and therefore build around them.
The best way to do this, besides not taking the superficially optimal class that does not fulfill your requirements, is to offload common tasks onto class features/items that do not necessarily require a higher level to function. In the case of the theurge, careful spell selection will mitigate the one level dip into an arcane caster class due to various feats neatly. There exists sufficient variety in spells that the trade off of one divine caster level for arcane casting is easily balanced.
Will this create an 'optimal blaster?" It depends on your requirements.
By enchanting magic items with common healing spells, it then becomes less necessary for you to have a maximal number of healing/buffing slots available. By finding an attack method that is not tied to caster-level during these first few critical levels, the need for higher-level spell slots is reduced.
Therefore, the "optimal" strategy is to take a level of ur-priest and theurge from that. Barring that, precocious apprentice or earth spell will provide a much shorter entry into mystic theurge, making it a slightly less non-optimal choice to take. At the end of the day, this is an area that has been well researched with much literature. Searching the literature for your given set of requirements will show solutions that you can test against your requirements level by level.
Be prepared to spend significant amounts of time on practical optimization, especially in systems that do not lend themselves to trivial computational modelling.
You are largely correct.
When they introduced haunts in Rise of the Runelords, they were still somewhat supernatural and ill-defined, and were super fun and creepy. We all still fondly remember Foxglove Manor. As much of the 3.5/PF community is intolerant of anything that's not entirely mechanistic, however, when they re-did haunts for Pathfinder (RotR was 3.5e originally) they did them in extra lame form as basically a trap. The only way you'd know any of those things to get rid of them is usually to make an artificial Knowledge check, which makes no sense you'd somehow know that, but that's where it is. Now, the haunts we hit in Carrion Crown recently were a lot more boring. "Oh let me channel energy again! Wait, no cleric? They're invulnerable!"
When I run haunts, I do two things. One, I throw out the trap/CR/etc crap and treat it more organically like a real inexplicable haunted-house thing. Then, the way to get rid of it is to research, ask creepy old Varisian ladies, ask the gods... And allow things to affect it that aren't just a cleric channeling energy, but symbolically appropriate stuff that any PC can do. Of course this means that haunts need to be "big deals" and not side throwaway encounters. Rather than have each individual haunt be its "own trap" I just make the whole haunt basically one thing, where you'd need to dispense with the whole thing and not "the scary painting in area A8." Haunts can be super fun and engaging if you reject the whole "it's a CR appropriate trap" approach and think "I'm directing Poltergeist."
Best Answer
That is to keep the magic items, in general, balanced.
A lot of spells scale better than Cure Light Wounds (which, in fact, scales terribly) and if brewing higher caster level items of lower level spells would make those potions more desirable than actually using high level spells, there would be no reason to make items using high level spells.
I will give you an example. A wand of snowball at CL 1 costs 750 gp and does 1d6 damage as a ranged touch attack. A wand of scorching ray at CL 3 costs 4500 gp and does 4d6 damage, also as ranged touch. If we take that snowball wand and increase it to CL 4, it will cost only 3000 gp to deal 4d6 damage, much cheaper than the wand of scorching ray.
This scenario can be seen on many magic items, not only wands. But if you make that comparison with cure wounds spells, it will show you that multiple castings of cure light wounds will net you a much better hit point per gp healing value than any higher version of the spell.
Example, a CL 1 cure light wounds potion costs 50 gp and heals 2-9 hp (average 5.5), a CL 3 cure moderate wounds potion costs 300 gp (6 times more) and heals 5-19 (average 12). While a potion of CL 3 cure light wounds would cost 150 gp and heal 1d8+3 (average 7,5). So two potions of cure light wounds will always heal more (2d8+6) than a single potion of cure moderate wounds (2d8+3), with the downside of requiring more time to heal that same amount of hit points. And if you make the lowest CL cure light wounds potion (1d8+1, 50 gp) you will heal much more than both options (6d8+6), but requiring six times longer to heal.
Even if you brew the cheapest potions for 50 gp, it only cost you half of that value (25 gp), it will always be a profitable profession.
Please realize that you can always create magic items using the minimum caster level for that item, as described under Magic Item Creation: