feat(desktop): DeerFlow visual redesign + stream hang fix + intelligence client

DeerFlow frontend visual overhaul:
- Card-style input box (white rounded card, textarea top, actions bottom)
- Dropdown mode selector (闪速/思考/Pro/Ultra with icons+descriptions)
- Colored quick-action chips (小惊喜/写作/研究/收集/学习)
- Minimal top bar (title + token count + export)
- Warm gray color system (#faf9f6 bg, #f5f4f1 sidebar, #e8e6e1 border)
- DeerFlow-style sidebar (新对话/对话/智能体 nav)
- Reasoning block, tool call chain, task progress visualization
- Streaming text, model selector, suggestion chips components
- Resizable artifact panel with drag handle
- Virtualized message list for 100+ messages

Bug fixes:
- Stream hang: GatewayClient onclose code 1000 now calls onComplete
- WebView2 textarea border: CSS !important override for UA styles
- Gateway stream event handling (response/phase/tool_call types)

Intelligence client:
- Unified client with fallback drivers (compactor/heartbeat/identity/memory/reflection)
- Gateway API types and type conversions

desktop/src/lib/intelligence-client/fallback-compactor.ts

@@ -0,0 +1,61 @@
+/**
+ * Intelligence Layer - LocalStorage Compactor Fallback
+ *
+ * Provides rule-based compaction for browser/dev environment.
+ */
+
+import type { CompactableMessage, CompactionResult, CompactionCheck, CompactionConfig } from '../intelligence-backend';
+
+export const fallbackCompactor = {
+  async estimateTokens(text: string): Promise<number> {
+    // Simple heuristic: ~4 chars per token for English, ~1.5 for CJK
+    const cjkChars = (text.match(/[\u4e00-\u9fff\u3040-\u30ff]/g) ?? []).length;
+    const otherChars = text.length - cjkChars;
+    return Math.ceil(cjkChars * 1.5 + otherChars / 4);
+  },
+
+  async estimateMessagesTokens(messages: CompactableMessage[]): Promise<number> {
+    let total = 0;
+    for (const m of messages) {
+      total += await fallbackCompactor.estimateTokens(m.content);
+    }
+    return total;
+  },
+
+  async checkThreshold(
+    messages: CompactableMessage[],
+    config?: CompactionConfig
+  ): Promise<CompactionCheck> {
+    const threshold = config?.soft_threshold_tokens ?? 15000;
+    const currentTokens = await fallbackCompactor.estimateMessagesTokens(messages);
+
+    return {
+      should_compact: currentTokens >= threshold,
+      current_tokens: currentTokens,
+      threshold,
+      urgency: currentTokens >= (config?.hard_threshold_tokens ?? 20000) ? 'hard' :
+               currentTokens >= threshold ? 'soft' : 'none',
+    };
+  },
+
+  async compact(
+    messages: CompactableMessage[],
+    _agentId: string,
+    _conversationId?: string,
+    config?: CompactionConfig
+  ): Promise<CompactionResult> {
+    // Simple rule-based compaction: keep last N messages
+    const keepRecent = config?.keep_recent_messages ?? 10;
+    const retained = messages.slice(-keepRecent);
+
+    return {
+      compacted_messages: retained,
+      summary: `[Compacted ${messages.length - retained.length} earlier messages]`,
+      original_count: messages.length,
+      retained_count: retained.length,
+      flushed_memories: 0,
+      tokens_before_compaction: await fallbackCompactor.estimateMessagesTokens(messages),
+      tokens_after_compaction: await fallbackCompactor.estimateMessagesTokens(retained),
+    };
+  },
+};