Add a TextSplitter in LangChain to share the model of the embedding model

### Feature request

Have you ever though to add an API endpoint that can serve as well as TextSplitter ?
It would replace the need to load in memory the same model for the text Chunker and the Embedder

https://python.langchain.com/docs/modules/data_connection/document_transformers/split_by_token#sentencetransformers

### Motivation

Create a LangChain TextSplitter that is based on the model Tokenizer to chunk long documents

### Your contribution

I feel ignorant about the AI domain (lack of knowledge)

The issue names no files, tests, or entry points. Start by locating the existing embedding model, tokenizer, and serving APIs, then compare them with LangChain's SentenceTransformers-based splitter. Done means a tokenizer-backed TextSplitter can share the loaded model with the embedder, with tests covering chunking and model reuse.

Written by the indexing model from the issue text.