Return Cohere embeddings as lists of floats (langchain-ai#1394)

This PR fixes the types returned by Cohere embeddings. Currently, Cohere
client returns instances of `cohere.embeddings.Embeddings`. Since the
transport layer relies on JSON, some numbers might be represented as
ints, not floats, which happens quite often. While that doesn't seem to
be an issue, it breaks some pydantic models if they require strict
floats.

langchain/embeddings/cohere.py

@@ -64,7 +64,7 @@ def embed_documents(self, texts: List[str]) -> List[List[float]]:
         embeddings = self.client.embed(
             model=self.model, texts=texts, truncate=self.truncate
         ).embeddings
-        return embeddings
+        return [list(map(float, e)) for e in embeddings]
 
     def embed_query(self, text: str) -> List[float]:
         """Call out to Cohere's embedding endpoint.
@@ -78,4 +78,4 @@ def embed_query(self, text: str) -> List[float]:
         embedding = self.client.embed(
             model=self.model, texts=[text], truncate=self.truncate
         ).embeddings[0]
-        return embedding
+        return list(map(float, embedding))