Add support for PPVCtrl

[owlowlohh]

Model/Pipeline/Scheduler description

I recently came across an impressive work called PPVCtrl, a controllable video generation model. It leverages an auxiliary condition encoder to transform a text-to-video generation model into a customizable video generator, all without retraining the original generator. It's akin to ControlNet, but for video generation.

Prompt	Reference Image	Control Videos	PP-VCtrl-5B-T2V	PP-VCtrl-5B-I2V
Group of fishes swimming in aquarium.
A boat with a flag on it is sailing on the sea.

Prompt	Reference Image	Control Videos	PP-VCtrl-5B-T2V	PP-VCtrl-5B-I2V
A rider in a dark helmet and white breeches is atop a chestnut horse...
A dark gray Mini Cooper is parked on a city street...

Prompt	Reference Image	Pose Videos	PP-VCtrl-5B-I2V
A young man with curly hair and a red t-shirt featuring a white logo is seen in various states of motion...
A woman models an Adrianna Papell women's gown, featuring a sleeveless...

Open source status

• The model implementation is available.
• The model weights are available (Only relevant if addition is not a scheduler).

Provide useful links for the implementation

https://github.com/PaddlePaddle/PaddleMIX/tree/develop/ppdiffusers/examples/ppvctrl