feat(caption): let live captions use their own speech model

The speech engine now keeps an auxiliary model next to the dictation model
and transcribes with whichever the request names, reloading it once if the
engine restarted. Settings > STT gains a live-caption model so captions can
run on large-v3-turbo while dictation keeps its own model. The runtime
minimum rises to 1.7.0 because older engines would silently ignore the
model choice.

Suggestion paging moves to Up/Down: the page follows the selection and the
last item waits while more candidates are being generated. The Left/Right
page shortcuts are removed; they did nothing until a page had filled and
clash with Intel's display-rotation hotkeys.

packages/i18n/src/locales/zh.json

@@ -375,7 +375,6 @@
   "popup.suggestion.hintNext": "下一个",
   "popup.suggestion.hintDismiss": "关闭",
   "popup.suggestion.hintMove": "移动",
-  "popup.suggestion.hintPage": "翻页",
   "popup.suggestion.loading": "生成中…",
   "keybinding.ui.sectionInput": "输入建议",
   "keybinding.action.suggestionAccept": "接受建议",
@@ -454,10 +453,6 @@
   "popup.suggestion.hintGeneratingMore": "正在生成更多…(最多 {{max}} 条)",
   "keybinding.action.suggestionPrev": "上一条建议",
   "keybinding.action.suggestionPrev.desc": "移动到上一条候选。",
-  "keybinding.action.suggestionPageNext": "下一页",
-  "keybinding.action.suggestionPageNext.desc": "显示下一页建议(最多 12 条)。",
-  "keybinding.action.suggestionPagePrev": "上一页",
-  "keybinding.action.suggestionPagePrev.desc": "显示上一页建议。",
   "input.insights.tabs.graph": "图谱",
   "input.graph.description": "Your sentences are stored as nodes and their relations as edges (what follows what, which share terms) so suggestions can pull personal context. Local only.",
   "input.graph.nodes": "句子节点",
@@ -524,5 +519,8 @@
   "popup.caption.waiting": "正在聆听… 第一条字幕可能需要几秒钟",
   "popup.caption.dragHint": "拖动以移动 · 双击复位",
   "settings.captionRefine": "按上下文润色字幕",
-  "settings.captionRefine.desc": "本地 AI 会根据上下文修正已确定字幕的空格、标点和听错的词。"
+  "settings.captionRefine.desc": "本地 AI 会根据上下文修正已确定字幕的空格、标点和听错的词。",
+  "settings.captionModel": "实时字幕模型",
+  "settings.captionModel.same": "与听写相同",
+  "settings.captionModel.desc": "仅实时字幕使用单独的模型识别，会与听写模型一同加载。"
 }