[feat-add] Fastembed Embedding provider integration

[lucifertrj]

• Added FastEmbed embedding provider under RAG support for local embedding models for faster processing.
• Included tests as well with the model_name="sentence-transformers/all-MiniLM-L6-v2" as used in qdrant/constants.py: DEFAULT_EMBEDDING_MODEL
• Included fastembed as supported provider under lib/crewai-tools - rag embedding_service.py

The provider is also tested with the following script:

from crewai.rag.embeddings.factory import build_embedder
from crewai.rag.embeddings.providers.fastembed.types import FastEmbedProviderSpec

embedding_model: FastEmbedProviderSpec = {
    "provider": "fastembed",
    "config": {
        "model_name": "sentence-transformers/all-MiniLM-L6-v2",
        "batch_size": 2},
}

embedder = build_embedder(embedding_model)
embeddings = embedder(["CrewAI supports FastEmbed embeddings."])
print(f"Embeddings: {len(embeddings)}")
print(f"Dimension: {len(embeddings[0])}")

GOAL: The reason for adding this provider is that Qdrant, under RAG, uses FastEmbed in config.py. It would be ideal to include FastEmbed as an embedding provider so it can also be used seamlessly with CrewAI tools.