I don't use AF much with my 50mm f/1.4, so I don't have any direct comparisons. However, I have read of issues with the 50mm AF being erratic at times. You are always going to have a greater risk of focusing issues with low contrast subjects, simply because the AF uses the contrast to make the decision on where to focus, however, it will only focus within the area of the active focus point(s), so any high contrast areas away from the focus point will be irrelevant. I once did a test using the 7D and 50mm f/1.4 at night with street lighting against my white garage door and surrounding brick work. I found that the focusing accuracy varied according to the ISO setting chosen, although it wasn't really a proper test (I was handholding for a start), so I could have been mistaken. If you only sent the lens for calibration without the body, then they won't necessarily be matched, so you could try microadjustment, but I would rule out othe causes first, in case it makes it worse. Try a similar subject in different lighting and two different subjects with different contrast levels in the same place, to see if it makes a difference. If it does, then MA is unlikely to help.
In your first example at the wedding, it's difficult to say for sure if the leaves are in front or behind the people, also, from the little of her chin visible in the crop, it looks like it may be in sharper focus than her chest, so that could be due to focus and recompose if you were using the lens wide open, as the focal plane would be changing. When you focus while pointing the camera up or down, then hold it level, it will have the same type of effect as using a tilt and shift lens, where even one degree of tilt can make a difference (although you can't compare one degree of camera tilt to one degree of a TSE lens tilt).