Allow for batched models to work in trt pipelines

[finlay-hudson]

This PR is to allow for batch mode to work with the TRT models in both img2img and txt2img pipelines. It mimics batching in default pytorch method, whereby all images have to be of consistent sizes throughout the batch.

Speeds

Tested on 50 iterations, base model, txt2img pipeline

Batch Size	Total Time (s)	Time Per Image (s)
1	5.26	5.26
4	19.78	4.94
8	38.39	4.79

Tested on 37 iterations, base model, img2img pipeline

Batch Size	Total Time (s)	Time Per Image (s)
1	4.09	4.09
4	15.45	3.85
8	30.25	3.78

Tested on 37 iterations, refiner model, img2img pipeline

Batch Size	Total Time (s)	Time Per Image (s)
1	4.48	4.48
4	17.24	4.31
8	33.53	4.19

[height]

Link Height tasks by mentioning a task ID in the pull request title or commit messages, or description and comments with the keyword link (e.g. "Link T-123").

💡Tip: You can also use "Close T-X" to automatically close a task when the pull request is merged.