fix: add additional_drop_params at global litellm_settings level

ai/litellm-config.yaml

@@ -31,10 +31,12 @@ litellm_settings:
   set_verbose: true
   # Disable prompt caching features
   cache: false
-  # Force strip specific parameters
+  # Force strip specific parameters globally
   allowed_fails: 0
   # Modify params before sending to provider
   modify_params: true
+  # Drop prompt_cache_key globally for all models
+  additional_drop_params: ["prompt_cache_key"]
 
 router_settings:
   allowed_fails: 0