Update docstring in base.py in MilvusVectorStore adding COSINE as ava…

…ilable similarity metric (run-llama#16031)

Update docstring in base.py adding COSINE as similarity metric available

According to Milvus Documentation, **COSINE** as similarity metric is supported (Both Milvus and Milvus Lite) but in Llama-Index docs was missing.

[Link to Milvus official docs](https://milvus.io/docs/metric.md?tab=floating#Similarity-Metrics)

I've checked in the [code](https://github.com/run-llama/llama_index/blob/723c2533ed4b7b43b7d814c89af1838f0f1994c2/llama-index-integrations/vector_stores/llama-index-vector-stores-milvus/llama_index/vector_stores/milvus/base.py#L256) , so indeed COSINE is supported, so no more further changes are needed.

llama-index-integrations/vector_stores/llama-index-vector-stores-milvus/llama_index/vector_stores/milvus/base.py

@@ -112,7 +112,7 @@ class MilvusVectorStore(BasePydanticVectorStore):
         doc_id_field (str, optional): The name of the doc_id field for the collection,
             defaults to DEFAULT_DOC_ID_KEY.
         similarity_metric (str, optional): The similarity metric to use,
-            currently supports IP and L2.
+            currently supports IP, COSINE and L2.
         consistency_level (str, optional): Which consistency level to use for a newly
             created collection. Defaults to "Session".
         overwrite (bool, optional): Whether to overwrite existing collection with same