Making rocket engines by metal 3D printing has become popular. A rocket engine bell and combustion chamber has one big rigid part with lots of channels and voids inside. That's the ideal case for 3D printing. Much simpler than building the thing up by machining and welding together many individual parts.
Maybe the main problem is that you want different materials in different parts of the structure?
E.g., the inner wall of the cooling channel in the bell is better made of something weaker but more heat-conductive, as noted in TFA. And, different materials in the chamber, throat, and bell.
I suppose you could have multiple feeders, on an additive system. But there are various reasons to want to use laser sintering of powdered metal, instead. And bonding the different metals is its own challenge. You could vary the mix of powder dropped at different levels, and rate of motion of the laser to match. But the combustion chamber probably wants concentric layers.
Plenty of hard choices and problems to solve. Engineers earn their keep.
I remember reading in an article that without 3D printing they could simply not get the cooling channel geometry they needed on the Super Dracos on Dragon 2.
Without 3D printing the combustion chamber they would need to use other cooling channel geometry, makin ghr engine heavier, bigger and less efficient.
There are different problems with 3D printing than with classical subtractive manufacturing. E.g. you need to get the supporting powder out of long thin passages in the cooling chamber - you don't have such problem with milling machine approach.
