The orientation of axes is entirely arbitary and so is ruled by convention. For things like free-body diagrams, converting polar co-ords to cartesian and so on, right is X+ and down is Y-, sin gives you your X component and cosine gives you Y. Since what we're doing is converting polar co-ords to cartesian, I just followed the convention I was taught. For folk from other backgrounds, there can be any number of variations of X, Y, up, down, sin, cos and so on because as long as you're still using the rules of geometry it doesn't *really* matter what else you do. A picture is still a picture, no matter how it's oriented or where you put it on the wall. It's down to you to manage your equations so they're consistent - the numbers don't give a slippery fig!
My method is entirely arbitary as I say, but that's the convention I've always both used and seen. Confusion normally comes because onscreen Y is inverted (Y- is UP) and the origin is top-left, but the origin of a co-ord system is arbitary too. If I stray from convention I always pop a postit on my monitor with the axes and origin marked because it's all too easy to slip back to 0,0 at bottom left and so on. Screen co-ords are even more arbitary (from our perspective) because there's rarely a functional relationship between X and Y, it's just us saying "plonk a sprite on the screen here" so there's no *mathematical* need to have 0,0 on hard top-left. There's even an argument for having the top-left of the viewport safe area as 0,0 ;0)
Also, I use theta for the angle, but I'm just causing trouble now!
Regards,
Mike