First of all, plants do not live on photosynthesis; all this does is facilitate an endothermic reaction that takes CO2 and water and converts it to O2 and carbohydrates. Plants actually need O2 and sugars to create the energy of life through exothermic reactions (like all life on Earth); the difference is that they've evolved on a planet that initially had virtually zero O2 in the atmosphere so had to use the energy of the sun to generate their own.
Of course, on a toxic new earth, many of the complex minerals required to sustain life was in the mineral surface so it would make sense that they would evolve roots to keep them in a single favourable space that they could effectively 'mine' for the minerals necessary for growth et al.
The problem I have with sentient plants; photosynthesis is a very energy intensive exercise and is only likely to evolve in a world where it becomes the only source of O2 and the carbohydrates necessary to sustain life. Nervous systems also are quite energy intensive and actually use electrical 'current' to drive and power them.
IMHO, this is the primary reason why no plant has ever developed even a rudimentary nervous system; it couldn't provide any tangible benefit to a life form that's stationary and can't react to an adverse environment in any other manner than dying. Photosynthesis on the other hand does add value and if you can only put energy into one of them, the latter is the one to choose.
Intelligence takes even more energy again; a human's energy requirements break down to around a quarter of that energy feeding brain activity. This is why beanies are so effective at keeping people warm in the snow; a LOT of the heat we generate and give off is through the head.
The practical upshot of this is that a 'creature' that has developed photosynthesis AND intelligence will be massively energy intensive. That energy 'cost' would have to fall below the opportunity benefit, and on Earth at least, that would be rare.
On your Venus(ish) world however, perhaps there are conditions that could make this work. I'm thinking of a thick atmosphere that contains CO2 and a 'nutrient soup' of enzymes, proteins, water (humidity), etc. The nutrients mean that a plant could evolve to be mobile because no location is especially more beneficial, and movement has advantages in adapting to changes in environment. The photosynthesis creates just enough O2 and sugars for energy production, which gets used immediately by the plant. The remaining heat energy absorbed somehow goes directly into the plant in a manner similar to photovoltaic solar cells, effectively powering a nervous system outside the chemical processes used for metabolism. You still need a reason to invest in intelligence and social interaction. The only one that comes to mind in this environment would be sex.
Plants on Earth are effectively androgynous but imagine a scenario where your venus plants are androgynous but with a twist; they can't self-pollinate. This scenario precludes insects as a pollination method; that would be easier and therefore social interactions would not develop. In this case, we'll say that pollination can only occur through direct contact, 'flower to flower', so to speak.
These mobile plants now have a reason to seek each other out (procreation) and mate selection also becomes an issue. Ultimately, the benefits of cooperative behaviours manifest, and communication, social structure, possibly tool building all ensue.
Given the simplicity of the nervous system and intelligence likely to develop, I'm going to assume that we're not talking the development of eyes, mouths or other 'conventional' sensory organs. Let's assume that contact is purely tactile. Communication would therefore likely be tactile in nature, probably some form of chemical transference. This would initially be limited so simple instinctual and (possibly) emotional concepts, but full sentience would require either a larger range of sensory inputs, or a LOT more time. Either way, it would be VERY interesting to see this in action.