Fixup GenerativeRequestSession

Author: sjmonson

src/guidellm/request/session.py

@@ -5,7 +5,7 @@
 from guidellm.backend.response import ResponseSummary
 from guidellm.request.request import GenerationRequest
 
-__all__ = ["RequestSession"]
+__all__ = ["GenerativeRequestSession", "RequestSession"]
 
 # TODO: Replace with specific types that implement needed features
 RequestT = TypeVar("RequestT")
@@ -36,7 +36,7 @@ def __init__(self, prompts: list[GenerationRequest]) -> None:
         self.prompts = prompts
         self.responses: list[str] = []
 
-    def get_request(self) -> GenerationRequest:
+    def get_next_request(self) -> GenerationRequest:
         completed_responses = len(self.responses)
         base_request = self.prompts[completed_responses].model_copy()
         base_request.content = "".join(
@@ -53,6 +53,9 @@ def get_request(self) -> GenerationRequest:
 
         return base_request
 
+    def get_next_delay(self) -> float:
+        return 0.0
+
     def push_response(self, response: ResponseSummary) -> None:
         if len(self.responses) < len(self.prompts):
             if response.response_output_tokens is not None: