feat(typesafe): add `ModerationMiddleware` for input content moderation

### Submission checklist - [x] This is a feature request, not a bug report or usage question. - [x] I added a clear and descriptive title that summarizes the feature request. - [x] I used the GitHub search to find a similar feature request and didn't find it. - [x] I checked the LangChain documentation and API reference to see if this feature already exists. - [x] This is not related to the langchain-community package. ### Package (Required) - [ ] langchain - [ ] langchain-openai - [ ] langchain-anthropic - [ ] langchain-classic - [ ] 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-openrouter - [ ] langchain-perplexity - [ ] langchain-qdrant - [x] langchain-typesafe - [ ] langchain-xai - [ ] Other / not sure / general ### Feature Description Related: #40556 (`langchain-typesafe` middleware) Add `ModerationMiddleware` to classify incoming user content before an agent runs and refuse requests that violate an application's content policy. Applications should be able to configure the categories to check and how flagged requests are handled. ### Use Case Applications may need to moderate user input for content such as violence, sexual content, or political topics, depending on their own policies and scope. For example, a domain-specific assistant may need to decline political discussions while allowing ordinary questions within its intended domain. A reusable middleware would let developers apply these rules consistently before the agent generates a response or invokes tools, without implementing a custom moderation layer for each application. ### Proposed Solution Provide a `ModerationMiddleware` in `langchain-typesafe` that checks the latest user message once per turn in `before_agent`. Allow …