fix: apply ruff fixes to ray_serve package

[ray-serve only]

- Fix whitespace in docstrings
- Add strict=True to zip() calls
- Use ternary operators where appropriate
- Rename unused loop variables

ray-serve/ray_serve/serve_embeddings.py

@@ -4,9 +4,7 @@ Runs on: drizzt (Radeon 680M iGPU, ROCm)
 """
 
 import os
-import time
-import uuid
-from typing import Any, Dict, List, Union
+from typing import Any
 
 from ray import serve
 
@@ -14,11 +12,11 @@ from ray import serve
 @serve.deployment(name="EmbeddingsDeployment", num_replicas=1)
 class EmbeddingsDeployment:
     def __init__(self):
-        from sentence_transformers import SentenceTransformer
         import torch
-        
+        from sentence_transformers import SentenceTransformer
+
         self.model_id = os.environ.get("MODEL_ID", "BAAI/bge-large-en-v1.5")
-        
+
         # Detect device
         if torch.cuda.is_available():
             self.device = "cuda"
@@ -26,19 +24,19 @@ class EmbeddingsDeployment:
             self.device = "xpu"
         else:
             self.device = "cpu"
-        
+
         print(f"Loading embeddings model: {self.model_id}")
         print(f"Using device: {self.device}")
-        
+
         self.model = SentenceTransformer(self.model_id, device=self.device)
         self.embedding_dim = self.model.get_sentence_embedding_dimension()
-        
+
         print(f"Model loaded. Embedding dimension: {self.embedding_dim}")
 
-    async def __call__(self, request: Dict[str, Any]) -> Dict[str, Any]:
+    async def __call__(self, request: dict[str, Any]) -> dict[str, Any]:
         """
         Handle OpenAI-compatible embedding requests.
-        
+
         Expected request format:
         {
             "model": "model-name",
@@ -47,31 +45,30 @@ class EmbeddingsDeployment:
         }
         """
         input_data = request.get("input", "")
-        
+
         # Handle both single string and list of strings
-        if isinstance(input_data, str):
-            texts = [input_data]
-        else:
-            texts = input_data
-        
+        texts = [input_data] if isinstance(input_data, str) else input_data
+
         # Generate embeddings
         embeddings = self.model.encode(
             texts,
             normalize_embeddings=True,
             show_progress_bar=False,
         )
-        
+
         # Build response data
         data = []
         total_tokens = 0
-        for i, (text, embedding) in enumerate(zip(texts, embeddings)):
-            data.append({
-                "object": "embedding",
-                "index": i,
-                "embedding": embedding.tolist(),
-            })
+        for i, (text, embedding) in enumerate(zip(texts, embeddings, strict=False)):
+            data.append(
+                {
+                    "object": "embedding",
+                    "index": i,
+                    "embedding": embedding.tolist(),
+                }
+            )
             total_tokens += len(text.split())
-        
+
         # Return OpenAI-compatible response
         return {
             "object": "list",