Text Tokenizer - GCS Bucket Permission Denied

Hi, I'm working with `videoprism_lvt_public_v1_base` like so:

```
from videoprism import models as vp
vp.load_text_tokenizer(tokenizer_name)
```

However, this gives me a permission denied error as the tokenizer is now hosted in a now-private GCS bucket: gs://t5-data/vocabs/cc_en.32000/sentencepiece.model.

I've also posted an issue here: https://github.com/google/sentencepiece/issues/1191, but any other help would be much appreciated. Thanks!

Start at `vp.load_text_tokenizer(tokenizer_name)` and investigate how it accesses `gs://t5-data/vocabs/cc_en.32000/sentencepiece.model`. Reproduce the permission-denied failure and determine whether the issue is in the repository's tokenizer loading path or access to the now-private bucket; done means the tokenizer can be loaded or the required access limitation is documented.

Written by the indexing model from the issue text.