Support evaluation and guardrails LM providers

[rafaelsandroni]

Checked

• I searched existing ideas and did not find a similar one
• I added a very descriptive title
• I've clearly described the feature request and motivation for it

Feature request

After facing a couple of hallucinations problems implementing AI agents, we decided to implement some language models to filter and evaluate the content between the app and llm provider. Where could we integrate it into langchain? Are any folder in the community integrations that are appropriate for it?

Motivation

Usually it's a common problem for large companies using langchain to deploy gen ai apps.

Proposal (If applicable)

Integrates it with evaluations and guardrails providers (e.g., https://metatext.ai/docs)