fix(agent): honor explicit thinking off (#2898)

* fix(agent): honor explicit thinking off

* fix(agent): address thinking off lint failures

* Clarify unset thinking level display

* fix ci

web/frontend/src/components/models/add-model-sheet.tsx

@@ -720,7 +720,7 @@ export function AddModelSheet({
                   <Input
                     value={form.thinkingLevel}
                     onChange={setField("thinkingLevel")}
-                    placeholder="off"
+                    placeholder={t("models.field.providerDefault")}
                   />
                 </Field>
 

web/frontend/src/components/models/edit-model-sheet.tsx

@@ -690,7 +690,7 @@ export function EditModelSheet({
                   <Input
                     value={form.thinkingLevel}
                     onChange={setField("thinkingLevel")}
-                    placeholder="off"
+                    placeholder={t("models.field.providerDefault")}
                   />
                 </Field>
 

web/frontend/src/i18n/locales/en.json

@@ -313,7 +313,8 @@
       "rpm": "Rate Limit (RPM)",
       "rpmHint": "Maximum requests per minute. 0 = no limit.",
       "thinkingLevel": "Thinking Level",
-      "thinkingLevelHint": "Extended thinking budget: off, low, medium, high, xhigh, adaptive.",
+      "thinkingLevelHint": "Leave blank to omit thinking_level and use the provider default. Values: off, low, medium, high, xhigh, adaptive.",
+      "providerDefault": "provider default",
       "maxTokensField": "Max Tokens Field",
       "maxTokensFieldHint": "Override the request field name for max tokens, e.g. max_completion_tokens.",
       "toolSchemaTransform": "Tool Schema Transform",

web/frontend/src/i18n/locales/pt-br.json

@@ -312,7 +312,8 @@
       "rpm": "Limite de Taxa (RPM)",
       "rpmHint": "Máximo de requisições por minuto. 0 = sem limite.",
       "thinkingLevel": "Nível de Pensamento",
-      "thinkingLevelHint": "Orçamento de pensamento estendido: off, low, medium, high, xhigh, adaptive.",
+      "thinkingLevelHint": "Deixe em branco para omitir thinking_level e usar o padrão do provider. Valores: off, low, medium, high, xhigh, adaptive.",
+      "providerDefault": "padrão do provider",
       "maxTokensField": "Campo de Max Tokens",
       "maxTokensFieldHint": "Sobrescreve o nome do campo de max tokens na requisição, ex: max_completion_tokens.",
       "toolSchemaTransform": "Transformação de Schema de Ferramentas",

web/frontend/src/i18n/locales/zh.json

@@ -313,7 +313,8 @@
       "rpm": "速率限制（RPM）",
       "rpmHint": "每分钟最大请求数，0 表示不限制。",
       "thinkingLevel": "思考级别",
-      "thinkingLevelHint": "扩展思考预算：off、low、medium、high、xhigh、adaptive。",
+      "thinkingLevelHint": "留空则不传 thinking_level，使用 Provider 默认值。可选值：off、low、medium、high、xhigh、adaptive。",
+      "providerDefault": "Provider 默认",
       "maxTokensField": "Max Tokens 字段名",
       "maxTokensFieldHint": "覆盖请求中 max_tokens 的字段名，例如 max_completion_tokens。",
       "toolSchemaTransform": "工具 Schema 转换",