Fix setting model device for CUDA

Previously, trying to run the model when ```device``` is set to ```cuda:0``` gave the following error:

```
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument index in method wrapper_CUDA__index_select)
```

This was happening because ```device``` was only being used to set the input's device while the model's device was being left as CPU. The fix is to also set the model's device using ```device```

# Screenshots

Before Changes [CUDA] (broken):

After Changes [CUDA] (fixed):

After Changes [CPU] (still works):

app.py

@@ -2,9 +2,11 @@ import gradio as gr
 from transformers import AutoTokenizer, AutoModelForSeq2SeqLM, pipeline
 import torch
 
-model = AutoModelForSeq2SeqLM.from_pretrained("Jayyydyyy/m2m100_418m_tokipona")
-tokenizer = AutoTokenizer.from_pretrained("facebook/m2m100_418M")
 device = "cuda:0" if torch.cuda.is_available() else "cpu"
+
+model = AutoModelForSeq2SeqLM.from_pretrained("Jayyydyyy/m2m100_418m_tokipona").to(device)
+tokenizer = AutoTokenizer.from_pretrained("facebook/m2m100_418M")
+
 LANG_CODES = {
     "English":"en",
     "toki pona":"tl"