The big reason for going to per-planet turrets was to eliminate the one big chokepoint.
And guess what... that failed. It's still the best defense strategy. The AI still has to be strong enough to beat that strong point... and will always be until all defenses "ressources" are limited by planets and not per galaxy. "Thankfully" some of the per-galaxy capped defenses are grossly overpowered - enabling multiple worlds to be protected. But, let's face it... there is absolutely no way that a chokepoint will ever become the worst strategy, ever. That's kind of the definition of a chokepoint.
If you had said that the chokepoint's power is comparatively "lower" compared to "regular planet" defenses before that change, I'd agree... yet the problem would also have been solved by reducing the quantity & power of per-galaxy defenses, or having the AI spread it's attacks more often. Or having the AI multiply its attacks when fighting a single choke point (that has been done if I remember well). So basically it never was "the solution" to accomplish that goal in the first place... just "one" amongst many possibilities.
A chokepoint will always be no worse than distributed defenses. Unless you do crazy, "Force the player to behave MY WAY! RAR!" penalties/AI bonuses. Consider, for a moment: The AI, as you suggested, currently gets a wave size bonus multiplier based on how many human systems can receive waves. At 1 system or less, it gets about a 50% bonus, and 7 or so and higher it gets a small penalty (10%? 20%? Something like that). But even under those circumstances, a single chokepoint is still better. 2 systems with 50% of your defenses each is worse versus the AI at 100% strength than 1 system receiving 200% strength AI waves. Actually, since unit count is a super-linear advantage, it's actually even better than just that. And if you do put in a crazy rule, like "the AI gets 100 times the strength against chokepoints!" you're basically saying "Players aren't allowed to play that way" - and that's not fun.
Spreading attacks out would have been OK, but only if it would have happened 100% of the time. If the AI was ever allowed to concentrate its forces, one of your partially defended planets would have instantly lost. Exowaves and CPAs come to mind. So, even then, you would have needed a single chokepoint.
There was talk of a small number of "Turret Enhancers" or "Extra turret controllers" to allow a small number of systems to get better defenses - but it resolved to either chokepoints (if the number was too low) or per-planet caps, if the number was high.
Back in the per-galaxy cap days, capturing Fac IVs or Fabricators was mostly throw-away, because the first Exowave or CPA to come along would destroy it. Unless it was behind your chokepoint.
There were many ideas that people came up with back when this change was being debated. They generally fell into still needing a chokepoint, because of Exowaves or CPAs, or being complicated but no better than what we have now.
The change that was made was the simplest, yet most comprehensive, one that was debated. We'd already had the experience of per-planet turrets introduced when the Core Turrets were introduced, and many player took a liking to it. I used to be a Champion of the Chokepoint, until I had the opportunity to play to win at high difficulties WITHOUT using a chokepoint. That right there was enough to make me change my mind.
However, I try to find what was lost / gained, I find the following:
- simplified beachhead made SF and threat mostly irrelevant.
- higher defence on all planets made it so threat delays more their attacks compared to before
- higher defense on chokepoint means the AI has to be made EVEN larger than before.
- higher micro-management.
I disagree that SF and Threat are irrelevant. Threat won't enter a planet unless it has the strength to defeat any human units present - and that includes beachhead turrets on AI worlds. Once they do get strong enough, and they will, your beachhead goes bye-bye. Then usually an occupied world or two, if the Threat is that strong. CPAs do an excellent job of cleaning the beaches in most games I play.
Exowaves are still nasty, and easily get tough enough to break through my fully defended systems. So I now need to do a defense in depth. I expect the Exowave or CPA to destroy my first system. That's why I also have lots of turrets in my second system. Sometimes, that system will get destroyed, too. Even once I've fought off an attack like that, it is still necessary to re-capture and rebuild those systems. If I can't do it before the next Exo or CPA, then I'm doomed.
Threat in general got a lot smarter over the past few versions. It can't be baited, it doesn't sit on wormholes, it doesn't scatter - instead it actively gathers, hides, and waits for a weakness. Even though it hasn't gotten stronger, it has gotten more dangerous - independent of the turret changes.
None of that is particularly fun to me past seeing hybrids suiciding on beachheads during one hour or two.
Ok, yeah, Hybrids are pretty doomed these days. They aren't smart enough to protect themselves, and will suicide against beachheads or other turrets. They need some serious TLC. I would be thrilled if the next expansion was nothing but Hybrids (and Astro Trains).
Then again, matter of taste ? And the AI is now getting ludicrous amount of ships at higher level which matter little because the efficiency of defending is so high. Which again, forced the game design to put "low" amount of scrap for the recycling and created a bunch of other issues (lag, carriers quantity, high number of starship, problem with caps of "parasited" units...). Would the other solutions have those effects ? Probably would have had other side effects, yeah. Still I think that a better solution than this was just waiting to be found.
About the rest of your post... 
I'm sorry, but there is no link in that change and the consequences you are speaking of.
- Energy management could have been tweaked to matter in other ways.
- Game difficulty is a direct consequence of giving more spawns / mechanics to the AI. The amount of defense that the player's got is irrelevant - if you give the AI 5 times what the player can handle, he'll lose, whether the player has 10 or 100 turrets.
- as said above, relative difficulty of defending "standard" worlds compared to "chokepoint" worlds could have been solved in different ways.
Energy could have been handled a different way than making the player spend it on turrets.
Resource income could have been handled a different way than salvage.
Increasing the risk of failed attacks could have been handled a different way than reprisals.
Increasing the AI strength vs chokepoints could have been handled in a different way.
But I don't know of any other way to have made playing on a Honeycomb or Lattice map viable.
But I don't know of any other way to have made protecting captured uniques viable on all map types.
So, while there were other things that could have been done, there was nothing a simple that hit almost all the points necessary.
So... I do understand why it was chosen - and why it pleases some, but I'd have liked a solution based on "'let's rebalance stuff" instead of "MOAR STUFF".
It would be nice if we could get another long series of rebalancings. Not just defenses, but AI strength, costs, etc. Last time we went through this, there were many ideas suggested that just couldn't be worked out.
If you have any ideas, I know I would like to discuss them. Something to tone down the general power curve that has come to both the human and the AI over the past few years could be very interesting.
