Mark the post-tool status handoff so silent reasoning never shows a stale tool status

After a tool round, the next model turn can reason server-side for tens
of seconds without streaming a single event. The footer kept the last
tool's status counting up ('Reading inventory. (40s)') while the model
was already composing the answer — observed live on the OpenRouter
reasoning route. OpenCode always shows a thinking state here; Pulse now
emits a model_processing workflow state at the start of every post-tool
turn ('Working on the response with the gathered results.'), which
model_thinking upgrades when reasoning deltas actually arrive.

Pinned by a new interaction-quality corpus scenario.

docs/release-control/v6/internal/subsystems/ai-runtime.md

@@ -2388,6 +2388,14 @@ deriving an older display status from `workflowStatusHistory`.
 
 ## Current State
 
+Post-tool model turns mark the status handoff: every agentic turn after the
+first emits a `model_processing` workflow state ("Working on the response
+with the gathered results.") before the provider call, so a provider that
+reasons server-side without streaming events can never leave the previous
+tool's status stale on screen; `model_thinking` still upgrades the status
+when reasoning deltas actually arrive. Pinned by the interaction-quality
+corpus scenario "post-tool model turn marks the status handoff".
+
 Session titles are model-generated after the first exchange
 (`internal/ai/chat/session_title.go`): the chat runtime upgrades the
 auto-truncated first-prompt placeholder once the done event has reached the

internal/ai/chat/agentic.go

@@ -857,6 +857,16 @@ func (a *AgenticLoop) executeWithTools(ctx context.Context, sessionID string, me
 			req = requestSanitizer(req)
 		}
 
+		// Post-tool model turns can reason server-side for a long time without
+		// streaming a single event; without a fresh status the footer keeps
+		// the last tool's message ("Reading inventory. (40s)") while the model
+		// is already composing the answer. Mark the handoff explicitly. The
+		// model_thinking status still upgrades this when reasoning deltas
+		// actually arrive.
+		if turn > 0 {
+			emitWorkflowState(callback, "model_processing", "Working on the response with the gathered results.", sessionFSMState(a.sessionFSM), "")
+		}
+
 		maxProviderAttempts := 2
 		err := error(nil)
 		for attempt := 1; attempt <= maxProviderAttempts; attempt++ {

internal/ai/chat/interaction_scenario_corpus_test.go

@@ -137,6 +137,19 @@ func interactionScenarios() []interactionScenario {
 			answerMustNotContain: []string{"call_a1b2c3d4"},
 			streamMustContain:    []string{"pulse_query"},
 		},
+		{
+			name:    "post-tool model turn marks the status handoff",
+			promise: "after tools run, the footer status hands off from the last tool to the model composing the answer — a provider that reasons server-side without streaming events never leaves a stale tool status on screen",
+			prompt:  "summarize my inventory",
+			calls: []scriptedProviderCall{
+				providerQueryToolCall("call_e5f6a7b8"),
+				providerContentDone("Inventory summarized."),
+			},
+			maxTurns:          4,
+			orderedTypes:      []string{"session", "tool_start", "tool_end", "workflow_state", "content", "done"},
+			forbiddenTypes:    []string{"error"},
+			streamMustContain: []string{"model_processing", "Working on the response with the gathered results"},
+		},
 		{
 			name:    "no-narrative turn falls back to a clean operator sentence",
 			promise: "when the model runs tools but never writes an answer, the user reads one clean sentence — not raw JSON, tool-result dumps, or provider call ids",