perf: tool rendering in conversation page (#7937)

* fix(dashboard): route conversation history tool messages through ToolCallCard

When viewing conversation history, large tool outputs (e.g. a single
git log --stat producing tens of KB) caused the browser renderer to
freeze. Root cause: formattedMessages mapped every role (including
tool / system / _checkpoint) into user/bot bubbles, and bot plain
strings went through markstream-vue's MarkdownRender. Single 88KB
tool messages plus 88-of-them adding up to ~349KB of synchronous
markdown parsing was enough to block the main thread for 5+ seconds.

This patch:

- Indexes tool-role messages by tool_call_id
- Filters formattedMessages to user/assistant only — tool, system and
  _checkpoint roles no longer render as standalone bubbles
- Converts assistant.tool_calls (OpenAI shape, with tc.name/tc.arguments
  fallbacks) into the existing tool_call MessagePart, attaching the
  paired result so MessageList's ToolCallCard renders it (default
  collapsed, no longer feeds large strings into the markdown renderer)
- Drops empty placeholder plain parts when an assistant message only
  carries tool_calls
- Sets ts/finished_ts to 0 as a sentinel: ToolCallCard.toolCallDuration
  returns "" when startTime <= 0, suppressing a misleading "0ms"
  duration that would otherwise appear because conversation history
  has no real timing data

Behavior change: tool results are now embedded in their assistant's
ToolCallCard.result instead of appearing as separate bot bubbles.
This matches the main chat UI's behavior.

Fixes #7929
Refs #7372 #7456

* style(dashboard): use single scrollbar in conversation history preview

ToolCallCard's result/args panes have their own max-height + overflow,
which produced a nested scrollbar when nested inside the history
preview's already-scrollable .conversation-messages-container. Override
those constraints inside the preview only — the outer 500px-bounded
container already provides scroll bounds, so a single scrollbar feels
cleaner. The main chat UI is unaffected.

---------

Co-authored-by: wanger <wanger@example.com>

Author: 10knamesmore

dashboard/src/views/ConversationPage.vue

@@ -539,28 +539,54 @@ export default {
 
         // 将对话历史转换为 MessageList 组件期望的格式
         formattedMessages() {
-            return this.conversationHistory.map(msg => {
-                console.log('处理消息:', msg.role, msg.content);
-                
-                // 将消息内容转换为 MessagePart[] 格式
-                const messageParts = this.convertContentToMessageParts(msg.content);
-                
-                if (msg.role === 'user') {
-                    return {
-                        content: {
-                            type: 'user',
-                            message: messageParts
-                        }
-                    };
-                } else {
+            // 按 tool_call_id 索引 tool 角色消息的执行结果
+            const toolResultsById = {};
+            for (const msg of this.conversationHistory) {
+                if (msg.role === 'tool' && msg.tool_call_id) {
+                    toolResultsById[msg.tool_call_id] = msg.content;
+                }
+            }
+
+            return this.conversationHistory
+                // tool / system 等非聊天角色不直接渲染为气泡，避免大文本走 markdown 路径卡死页面
+                .filter(msg => msg.role === 'user' || msg.role === 'assistant')
+                .map(msg => {
+                    console.log('处理消息:', msg.role, msg.content);
+
+                    const messageParts = this.convertContentToMessageParts(msg.content)
+                        // 丢弃 convertContentToMessageParts 兜底插入的空 plain，避免 assistant 仅有工具调用时渲染空气泡
+                        .filter(part => part.type !== 'plain' || (part.text && part.text.trim()));
+
+                    // 把 OpenAI 风格的 assistant.tool_calls 转成 MessageList 已支持的 tool_call part
+                    if (msg.role === 'assistant' && Array.isArray(msg.tool_calls) && msg.tool_calls.length) {
+                        const toolCalls = msg.tool_calls.map(tc => {
+                            const fn = tc.function || {};
+                            return {
+                                id: tc.id,
+                                name: fn.name || tc.name,
+                                args: fn.arguments ?? tc.arguments,
+                                result: toolResultsById[tc.id],
+                                // 历史回放无真实耗时数据：
+                                // ts: 0  → ToolCallCard.toolCallDuration 在 startTime<=0 时早退，跳过时长显示
+                                // finished_ts: 1 → MessageList.toolCallStatusText 视为已完成（避免误显示"运行中"）
+                                ts: 0,
+                                finished_ts: 1,
+                            };
+                        });
+                        messageParts.push({ type: 'tool_call', tool_calls: toolCalls });
+                    }
+
+                    const finalParts = messageParts.length
+                        ? messageParts
+                        : [{ type: 'plain', text: '' }];
+
                     return {
                         content: {
-                            type: 'bot',
-                            message: messageParts
+                            type: msg.role === 'user' ? 'user' : 'bot',
+                            message: finalParts,
                         }
                     };
-                }
-            });
+                });
         }
     },
 
@@ -1173,6 +1199,18 @@ export default {
     background-color: #f9f9f9;
 }
 
+/* 让 ToolCallCard 内部的 args/result 自然展开，由外层容器统一滚动，避免双滚动条 */
+.conversation-messages-container .detail-json,
+.conversation-messages-container .detail-result {
+    max-height: none;
+    overflow: visible;
+}
+
+/* 历史回放无真实状态数据，隐藏 IPython 工具的"已完成"标签，与其它工具卡片保持一致 */
+.conversation-messages-container .tool-call-inline-status {
+    display: none;
+}
+
 /* 暗色模式下的聊天消息容器 */
 .v-theme--dark .conversation-messages-container {
     background-color: #1e1e1e;