text-splitters: remove `sentence-transformers` dependency

### Privileged issue - [x] I am a LangChain maintainer, or was asked directly by a LangChain maintainer to create an issue here. ### Issue Content Search the entire repo and refactor any implementation that depends on the `sentence-transformers` library, effectively removing the sub-dependencies of `pillow` and `torch` <details> <summary>Agent Context</summary> <open-swe-do-not-edit-task-plan> { "tasks": [ { "id": "c26efe20-a2c4-4a0f-a8ca-b176996e8d38", "taskIndex": 0, "request": "[original issue]\n**refactor: remove `sentence-transformers` dependency**\n### Privileged issue\n\n- [x] I am a LangChain maintainer, or was asked directly by a LangChain maintainer to create an issue here.\n\n### Issue Content\n\nSearch the entire repo and refactor any implementation that depends on the `sentence-transformers` library, effectively removing the sub-dependencies of `pillow` and `torch`", "title": "Remove sentence-transformers dependency from LangChain", "createdAt": 1753920572289, "completed": true, "planRevisions": [ { "revisionIndex": 0, "plans": [ { "index": 0, "plan": "**Refactor SentenceTransformersTokenTextSplitter to use transformers library directly**\n- Modify `/home/daytona/langchain/libs/text-splitters/langchain_text_splitters/sentence_transformers.py`\n- Replace `from sentence_transformers import SentenceTransformer` with imports from `transformers` library\n- Load the model using `AutoModel.from_pretrained()` and tokenizer using `AutoTokenizer.from_pretrained()`\n- Update the class to use the tokenizer directly instead of accessing it through `self._model.tokenizer`\n- Preserve the same public API and functionality including `max_seq_length` property access", "completed": true, "summary": "Successfully refactored SentenceTransformersTokenTextSplitter to use the transformers library directly:\n\n1. **Replaced imports**: …