fix: remove chunk string parameter from ResponseChunkProcessor interface

Plugins read the chunk via response.CurrentChunk and mutate via
response.SetChunk(). The separate chunk string parameter created
ambiguity for implementors.

Signed-off-by: Noy Itzikowitz <nitzikow@redhat.com>

Author: noyitz

pkg/framework/interface/requesthandling/plugins.go

@@ -52,12 +52,12 @@ type ResponseProcessor interface {
 }
 
 // ResponseChunkProcessor processes individual response body chunks as they
-// stream through without buffering. The framework converts the raw chunk bytes
-// to a string once and passes it to all chunk processors. Plugins receive the
-// InferenceResponse to allow header mutation.
+// stream through without buffering. The framework sets response.CurrentChunk
+// before calling plugins. Plugins read the chunk via response.CurrentChunk
+// and mutate it via response.SetChunk().
 type ResponseChunkProcessor interface {
 	plugin.Plugin
-	ProcessResponseChunk(ctx context.Context, cycleState *plugin.CycleState, response *InferenceResponse, chunk string, isFinal bool) error
+	ProcessResponseChunk(ctx context.Context, cycleState *plugin.CycleState, response *InferenceResponse, isFinal bool) error
 }
 
 type PostProcessor interface {

pkg/handlers/response.go

@@ -169,7 +169,7 @@ func (s *Server) runResponseChunkProcessors(ctx context.Context, cycleState *plu
 			verboseLogger.Info("Executing response chunk plugin", "plugin", cp.TypedName())
 		}
 		before := time.Now()
-		err := cp.ProcessResponseChunk(ctx, cycleState, response, response.CurrentChunk, isFinal)
+		err := cp.ProcessResponseChunk(ctx, cycleState, response, isFinal)
 		metrics.RecordPluginProcessingLatency(responsePluginExtensionPoint, cp.TypedName().Type, cp.TypedName().Name, time.Since(before))
 		if err != nil {
 			return err