v0.20.0: thinking blocks, chat windows overhaul, scroll fix

Backend:
- Add isThinking flag to agentEvent for text before tool calls
- Separate thinking from response text in runChatTurn and continue.go
- Persist thinking in a dedicated field in message content

Frontend:
- Add thinking field to MessageContent, ChatMessage, ChatTextEvent types
- Create ThinkingBlock.svelte — collapsible block with brain icon
- SSE handler moves text_delta content to thinking on isThinking flag
- Render thinking block between tools and response in ChatThread
- Fix chat window scroll reset on focus change (stable windowKeys order)
- Remove redundant #key id wrapper in WindowLayer
- Enlarge sidebar rail (24→32 default, 40→60 max)
- Remove glyph from sidebar, square graph at top
- Replace AgentTrace/ToolCallCard/UnifiedTimeline with TurnTrace/ToolLine

web/src/lib/stores/chat.ts

@@ -25,6 +25,7 @@ export interface ChatMessage {
   id: string
   role: 'user' | 'assistant'
   text: string
+  thinking?: string
   tools: ToolCallResult[]
   pendingApprovals: PendingApproval[]
   created_at?: string
@@ -223,6 +224,7 @@ function toChatMessages(msgs: Message[]): ChatMessage[] {
       id: m.id,
       role: m.role as 'user' | 'assistant',
       text: content?.text ?? '',
+      thinking: content?.thinking ?? undefined,
       tools,
       pendingApprovals: extractApprovals(tools),
       created_at: m.created_at
@@ -230,6 +232,16 @@ function toChatMessages(msgs: Message[]): ChatMessage[] {
   })
 }
 
+function chatMessagesChanged(a: ChatMessage[], b: ChatMessage[]): boolean {
+  if (a.length !== b.length) return true
+  for (let i = 0; i < a.length; i++) {
+    if (a[i].id !== b[i].id || a[i].role !== b[i].role || a[i].text !== b[i].text || a[i].tools.length !== b[i].tools.length) {
+      return true
+    }
+  }
+  return false
+}
+
 export async function loadSessionMessages(sessionId: string) {
   currentSession.set(sessionId)
   // This is a fresh view of sessionId's current (REST-loaded) state — reset
@@ -270,15 +282,11 @@ function startPolling(sessionId: string) {
     if (pollingSessionId !== sessionId || get(currentSession) !== sessionId) return
     const msgs = await fetchMessages(sessionId)
     if (get(streaming) || pollingSessionId !== sessionId) return // re-check: the fetch itself takes time
-    // No cheap "anything new?" check: the auto-continuation worker updates a
-    // placeholder message IN PLACE as each tool call lands (see
-    // cmd/nomos/continue.go), so the message COUNT stays the same while the
-    // content changes — a length-only diff (the previous version of this
-    // code) never detected those updates and progress looked frozen even
-    // though the backend was actively working. Just re-set every tick;
-    // Svelte's own diffing keeps the actual re-render cheap.
-    sessionMessages.set(msgs)
-    messages.set(toChatMessages(msgs))
+      const incoming = toChatMessages(msgs)
+      if (chatMessagesChanged(get(messages), incoming)) {
+        sessionMessages.set(msgs)
+        messages.set(incoming)
+      }
   }, 3000)
 }
 
@@ -401,7 +409,15 @@ export function sendMessage(text: string) {
         messages.update((ms) => {
           const last = ms[ms.length - 1]
           if (last && last.role === 'assistant') {
-            ms[ms.length - 1] = { ...last, text: ev.data }
+            if (ev.is_thinking) {
+              ms[ms.length - 1] = {
+                ...last,
+                thinking: (last.thinking || '') + ev.data,
+                text: ''
+              }
+            } else {
+              ms[ms.length - 1] = { ...last, text: ev.data }
+            }
           }
           return [...ms]
         })
@@ -596,7 +612,10 @@ function startSessionPolling(sessionId: string) {
       if (get(chat.streaming) && get(chat.connectionState) === 'connected') return
       const msgs = await fetchMessages(sessionId)
       if (get(chat.streaming)) return // re-check: the fetch itself takes time
-      chat.messages.set(toChatMessages(msgs))
+      const incoming = toChatMessages(msgs)
+      if (chatMessagesChanged(get(chat.messages), incoming)) {
+        chat.messages.set(incoming)
+      }
       // F3 safety net: if we're recovering from a dropped SSE but the
       // session's task has already reached a turn-ended status, clear the
       // stuck disconnected/streaming flags. Catches the edge where the
@@ -728,7 +747,15 @@ export function sendSessionMessage(sessionId: string, text: string) {
         chat.messages.update((ms) => {
           const last = ms[ms.length - 1]
           if (last && last.role === 'assistant') {
-            ms[ms.length - 1] = { ...last, text: ev.data }
+            if (ev.is_thinking) {
+              ms[ms.length - 1] = {
+                ...last,
+                thinking: (last.thinking || '') + ev.data,
+                text: ''
+              }
+            } else {
+              ms[ms.length - 1] = { ...last, text: ev.data }
+            }
           }
           return [...ms]
         })
@@ -869,7 +896,15 @@ export function startTask(text: string, onSession: (sessionId: string) => void):
       c.messages.update((ms) => {
         const last = ms[ms.length - 1]
         if (last && last.role === 'assistant') {
-          ms[ms.length - 1] = { ...last, text: ev.data }
+          if (ev.is_thinking) {
+            ms[ms.length - 1] = {
+              ...last,
+              thinking: (last.thinking || '') + ev.data,
+              text: ''
+            }
+          } else {
+            ms[ms.length - 1] = { ...last, text: ev.data }
+          }
         }
         return [...ms]
       })