Memory leaks in plain LLM calls

### Checked other resources - [x] This is a bug, not a usage question. - [x] I added a clear and descriptive title that summarizes this issue. - [x] I used the GitHub search to find a similar question and didn't find it. - [x] I am sure that this is a bug in LangChain rather than my code. - [x] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). - [x] This is not related to the langchain-community package. - [x] I posted a self-contained, minimal, reproducible example. A maintainer can copy it and run it AS IS. ### Package (Required) - [ ] langchain - [x] langchain-openai - [ ] langchain-anthropic - [ ] langchain-classic - [x] langchain-core - [ ] langchain-model-profiles - [ ] langchain-tests - [ ] langchain-text-splitters - [ ] langchain-chroma - [ ] langchain-deepseek - [ ] langchain-exa - [ ] langchain-fireworks - [ ] langchain-groq - [ ] langchain-huggingface - [ ] langchain-mistralai - [ ] langchain-nomic - [ ] langchain-ollama - [ ] langchain-perplexity - [ ] langchain-prompty - [ ] langchain-qdrant - [ ] langchain-xai - [x] Other / not sure / general ### Related Issues / PRs Potentially related, not sure: - https://github.com/langchain-ai/langchain/issues/30667 - https://github.com/openai/openai-python/issues/2672#issuecomment-3380147001 - https://github.com/pydantic/pydantic/issues/12446 (merged but not released yet) ### Reproduction Steps / Example Code (Python) Download the attached zip file. It uses UV. ```bash export PYTHONMALLOC=malloc export OPENAI_API_KEY="".." export GEMINI_API_KEY=".." export GOOGLE_APPLICATION_CREDENTIALS=".." # generate memray file uv run memray run <python file> # generate flamegraph uv run memray flamegraph <memray file> --leaks ``` Below is a code for one of the attached scripts (v1): ```python from langchain_openai import ChatOpenAI system_prompt = "You are a helpful assistant" user_prompt = "What is 1.5 + 2.7? Provide just the final numerical answer." client…