docs: migrate embedder→embedding_model and require vectordb across tool docs; add provider examples (en/ko/pt-BR)

[tonykipkemboi]

Summary

• Replace config.embedder with config.embedding_model in docs.
• Mark vectordb as required; add commented examples for Chroma Settings and Qdrant VectorParams.
• Add provider-specific commented examples (OpenAI, Google, Cohere, Ollama).
• Update English, Korean, and Portuguese File & Document tool pages; minor KO/PT-BR Datadog tweaks.
• Aligns docs with current RagTool config parsing.

Motivation

• Tools now read embedding configuration via embedding_model and support vectordb configs directly; old embedder examples are misleading.

Scope of changes

• en: PDF/TXT/XML/MDX/DOCX/CSV/Directory
• ko: PDF/TXT/XML/MDX/DOCX/CSV/Directory
• pt-BR: PDF/TXT/XML/MDX/DOCX/Directory
• KO/PT-BR observability: Datadog minor image frame tweaks

Developer-facing migration

• Before:

 config.embedder = { provider, config: { model, ... } }

• After:

config.embedding_model = { provider, config: { model, ... } }
config.vectordb = { provider: "chromadb"|"qdrant", config: { settings? | vectors_config? } }

Minimal example

from chromadb.config import Settings
tool = PDFSearchTool(
  config={
    "embedding_model": {
      "provider": "openai",
      "config": {"model": "text-embedding-3-small"}
    },
    "vectordb": {
      "provider": "chromadb",
      "config": {
        # "settings": Settings(persist_directory="/content/chroma", allow_reset=True, is_persistent=True),
      }
    }
  }
)

Notes

• Collection name is managed by the adapter (default: "rag_tool_collection").
• OpenAI key auto-read from env; “model” is mapped to “model_name”.

[lucasgomide]