Because from a "worth it" viewpoint most of the ricing mods cost power and street manners in addition to dollars. It's a basic car. By definition, every design decision was made to optimize for reliability, fuel economy and production cost. There are no quick inexpensive fixes to reverse tens of man years of engineering focus.
Many of the ricer mods also have outlandish performance claims, and typical forum goers simply do additive math at the upper range when bench racing. Sure, a less restrictive airbox could add 10 hp *IF* it was the airflow bottleneck. On a stock engine many of these mods could wind up subtracting power. The above computer analogy would be more appropos if that bling ram and heatsink lowered the CPU frequency by a few hundred mhz.
And a LSD on a car without enough torque to lose traction in the first place is simply additional drag and added wear to the tires. It won't make you corner harder, in fact just the opposite.
Actually the ram thing is spot on. If you put in overclocking ram to run at stock it's probably going to be high latency, high frequency ram which means slightly less performance with stock settings. 9-9-9-18 @ 1333mhz will always do worse than 7-7-7-15 @ 1333mhz.
The same with the intake resonator. You want turbulence in your intake, but only just a boundary layer on the outside wall to decrease fluid friction.v Your resonator and intake should already be designed ot do that anyways. Removing the resonator does 2 things, it turns the interior airflow from laminar to turbulent flow and it decreases the resonant frequency effect which means that you not only get less airflow, you also decrease the overpressure delivered to the valves at the rpm that the resonator is designed for (which in a R18A1 is probably 3000-4500 rpm).
It does open up the airflow at higher rpms from less restriction, but probably not enough to counteract the last 2 problems which means you'll get a "swirling" intake sound, lower midrange power and most likely the same, or very, very slightly increased power that can also just be attributed to the standard deviation of error between dyno runs.
Resonator removal is almost always a bad idea unless you can model the fluid dynamics yourself and design/fabricate anew one for the targeted rpm range you want more power at (which in all likelihood would be 5000-6500rpm since that's really the only place it would be able to improve anything measurably, though not substantially).
The only other reason for resonator removal is in the case of forced induction or other major engine intake mods (like a new head), at which point you probably wouldn't have a stock intake anymore anyways.