fix(backend): restore response_mime_type=json, raise max_output_tokens to 16384

Regular generation was hitting MAX_TOKENS at 8192. Constrained decoding with
16384 should be a viable middle ground between the truncation at 8192 and the
timeout at 65536.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

backend/innercontext/api/products.py

@@ -359,7 +359,8 @@ def parse_product_text(data: ProductParseRequest) -> ProductParseResponse:
         contents=f"Extract product data from this text:\n\n{data.text}",
         config=genai_types.GenerateContentConfig(
             system_instruction=_product_parse_system_prompt(),
-            max_output_tokens=8192,
+            response_mime_type="application/json",
+            max_output_tokens=16384,
             temperature=0.0,
         ),
     )