Update README.md

Add comment of max_seq_len to sentence transformers, especially for efficient training:
# Optionally, lower the maximum sequence length for lower memory usage
# model.max_seq_length = 8192

README.md

@@ -79,6 +79,9 @@ model = SentenceTransformer("Qwen/Qwen3-Embedding-4B")
 #     tokenizer_kwargs={"padding_side": "left"},
 # )
 
+# Optionally, lower the maximum sequence length for lower memory usage
+# model.max_seq_length = 8192
+
 # The queries and documents to embed
 queries = [
     "What is the capital of China?",