Silence placeholder/continuing messages in the tool-call auto-retry path

The tool loop's max-iterations and empty-response fallbacks were showing
placeholder assistant bubbles ("[Tool loop reached maximum iterations]",
"[No response from the model — retrying]") followed by a "↩ Continuing…"
system message before silently re-running. None of that added anything
for the user, so the auto-continue now happens without any visible
message when there's no real content to show; genuine partial content
is still displayed as before, and usage/cost tracking is unaffected.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

oAI/ViewModels/ChatViewModel.swift

@@ -360,7 +360,6 @@ Don't narrate future actions ("Let me...") - just use the tools.
     }
 
     func startAutoContinue() {
-        showSystemMessage("↩ Continuing…")
         silentContinuePrompt = "Please continue from where you left off."
         Task { @MainActor in
             generateAIResponse(to: "", attachments: nil)
@@ -1475,9 +1474,8 @@ Don't narrate future actions ("Let me...") - just use the tools.
                         if finalContent.isEmpty {
                             // Some models (observed with Qwen via OpenRouter) stop calling tools
                             // after a long tool-call chain but return no summarizing text at all.
-                            // Surface a placeholder and nudge a follow-up turn instead of a silent blank bubble.
+                            // Silently nudge a follow-up turn instead of showing a placeholder bubble.
                             finishedWithEmptyContent = true
-                            finalContent = "[No response from the model — retrying]"
                         }
                         break
                     }
@@ -1566,9 +1564,7 @@ Don't narrate future actions ("Let me...") - just use the tools.
                     // If this was the last iteration, note it
                     if iteration == maxIterations - 1 {
                         hitIterationLimit = true  // We're exiting with pending tool calls
-                        finalContent = response.content.isEmpty
-                            ? "[Tool loop reached maximum iterations]"
-                            : response.content
+                        finalContent = response.content
                     }
                 }
 
@@ -1577,41 +1573,57 @@ Don't narrate future actions ("Let me...") - just use the tools.
                     wasCancelled = true
                 }
 
+                // If we hit the iteration limit or the model returned no text at all, silently
+                // nudge a follow-up turn instead of showing a placeholder/blank bubble.
+                let willAutoContinue = (hitIterationLimit || finishedWithEmptyContent) && !wasCancelled
+
                 // Display the final response as an assistant message
                 let responseTime = Date().timeIntervalSince(startTime)
-                let assistantMessage = Message(
-                    role: .assistant,
-                    content: finalContent,
-                    tokens: totalUsage?.completionTokens,
-                    cost: nil,
-                    timestamp: Date(),
-                    attachments: nil,
-                    responseTime: responseTime,
-                    wasInterrupted: wasCancelled,
-                    modelId: modelId,
-                    generatedImages: finalImages.isEmpty ? nil : finalImages
-                )
-                messages.append(assistantMessage)
-
-                // Calculate cost
-                if let usage = totalUsage, let model = selectedModel {
-                    let hasPricing = model.pricing.prompt > 0 || model.pricing.completion > 0
-                    let cost: Double? = hasPricing ? calculateCost(usage: usage, pricing: model.pricing) : nil
-                    if let index = messages.lastIndex(where: { $0.id == assistantMessage.id }) {
-                        messages[index].cost = cost
+                if willAutoContinue && finalContent.isEmpty {
+                    // Nothing worth showing yet — still record usage/cost for this turn.
+                    if let usage = totalUsage, let model = selectedModel {
+                        let hasPricing = model.pricing.prompt > 0 || model.pricing.completion > 0
+                        let cost: Double? = hasPricing ? calculateCost(usage: usage, pricing: model.pricing) : nil
+                        sessionStats.addMessage(
+                            inputTokens: usage.promptTokens,
+                            outputTokens: usage.completionTokens,
+                            cost: cost
+                        )
                     }
-                    sessionStats.addMessage(
-                        inputTokens: usage.promptTokens,
-                        outputTokens: usage.completionTokens,
-                        cost: cost
+                } else {
+                    let assistantMessage = Message(
+                        role: .assistant,
+                        content: finalContent,
+                        tokens: totalUsage?.completionTokens,
+                        cost: nil,
+                        timestamp: Date(),
+                        attachments: nil,
+                        responseTime: responseTime,
+                        wasInterrupted: wasCancelled,
+                        modelId: modelId,
+                        generatedImages: finalImages.isEmpty ? nil : finalImages
                     )
+                    messages.append(assistantMessage)
+
+                    // Calculate cost
+                    if let usage = totalUsage, let model = selectedModel {
+                        let hasPricing = model.pricing.prompt > 0 || model.pricing.completion > 0
+                        let cost: Double? = hasPricing ? calculateCost(usage: usage, pricing: model.pricing) : nil
+                        if let index = messages.lastIndex(where: { $0.id == assistantMessage.id }) {
+                            messages[index].cost = cost
+                        }
+                        sessionStats.addMessage(
+                            inputTokens: usage.promptTokens,
+                            outputTokens: usage.completionTokens,
+                            cost: cost
+                        )
+                    }
                 }
 
                 isGenerating = false
                 streamingTask = nil
 
-                // If we hit the iteration limit, or the model returned no text at all, nudge a follow-up turn
-                if (hitIterationLimit || finishedWithEmptyContent) && !wasCancelled {
+                if willAutoContinue {
                     startAutoContinue()
                 }