(This might also be a good question to post in the heat transfer/thermo forum...)
What matters most is that the water goes in the bottom port and comes out the top port. Any gases released from the water from bends, twists, velocity changes, and throttling will travel up with the stream naturally versus fighting a bulk stream traveling in a net downward direction.
The air velocity through a coil might be 500 feet per minute or higher. At this speed, if the two row coil is (for example) a foot wide, each air molecule would spend 0.12 seconds between the fins before bouncing around off the fins and leaving with increased kinetic energy. With this short a period of contact for heat transfer, I doubt if there would be a difference in the overall outlet conditions if the inlet portion of the one-foot section were hotter than the outlet portion, or vice versa.
So I say make sure the inlet is at the bottom and the outlet is at the top, and install it in the direction that allows the best maintenance access to the control valve, circuit setter, and strainer.
Evidence might refute this. But this seems sensible... What do you think? -CB