fix: structured output prompt with example + DDG timeout 4s

rag_chat_use_case.py:
- Add concrete EXAMPLE block to both prompts so LLM sees exactly what format to produce
- Shorter, cleaner rules section
- Both execute_chat and execute_stream use identical prompt

.env:
- LIVE_SEARCH_TIMEOUT: 2.0 → 4.0s (HF Space has slower network than local)
DDG was timing out at 1.5s causing 0 live results

.env

@@ -79,7 +79,7 @@ ACCESS_TOKEN_EXPIRE_MINUTES=60
 
 # --- Hybrid Search Settings ---
 ENABLE_HYBRID_SEARCH=true
-LIVE_SEARCH_TIMEOUT=2.0
+LIVE_SEARCH_TIMEOUT=4.0
 LIVE_SEARCH_MAX_RESULTS=5
 LIVE_SEARCH_WEIGHT=0.5
 DB_SEARCH_WEIGHT=0.5

src/core/use_cases/rag_chat_use_case.py

@@ -865,34 +865,44 @@ JSON:"""
 
         prompt = f"""You are ARKI AI, a real-time Ethiopia & Africa news assistant. Today: {datetime.utcnow().strftime("%B %d, %Y")}.
 
-SOURCE INDEX (cite by number — these are the ONLY sources you may use):
+SOURCE INDEX (the ONLY sources you may cite):
 {source_index_lines if source_index_lines else "NO SOURCES RETRIEVED."}
 
-STRICT RULES:
-- Use ONLY facts from the News Context below. NEVER use training data.
-- Cite every fact: [1], [2], etc. Only use numbers that exist in the Source Index above.
-- Non-English articles: translate to English in your answer.
-- Always respond in English.
+RULES:
+1. Use ONLY facts from the News Context. Never use training data.
+2. Cite every fact inline: [1], [2], etc.
+3. Translate non-English articles to English.
+4. Respond in English only.
+5. End with: FOLLOW_UP: q1 | q2 | q3
 
-OUTPUT FORMAT — use this exact structure:
+REQUIRED OUTPUT FORMAT (follow exactly):
 
-## [Short headline summarizing the main news]
+## [One-line headline]
 
-**[Topic 1]**
-- Key fact from source [N]
-- Key fact from source [N]
+**[Category name]**
+- [Fact]. [1]
+- [Fact]. [2]
 
-**[Topic 2]** (if applicable)
-- Key fact from source [N]
+**[Another category]** *(if applicable)*
+- [Fact]. [N]
 
-> 💡 *[One sentence summary of the overall situation]*
+> 💡 [One sentence overall summary]
 
-FOLLOW_UP: question1 | question2 | question3
+EXAMPLE of correct format:
+---
+## Ethiopia: Conflict & Economy Updates
 
-EVALUATION GUIDE:
-- If sources directly answer the question → use the format above
-- If sources are related but not exact → start with "I found related news:" then use the format
-- If no relevant sources → say "I couldn't find relevant news on that topic in today's feed." and STOP
+**Ethiopia-Sudan Tensions**
+- Sudan recalled its ambassador after accusing Ethiopia of drone strikes on Khartoum airport. [2]
+- Ethiopia rejected the accusations as "baseless" and blamed external parties. [3]
+
+**Economic Developments**
+- Ethiopia's GERD dam was inaugurated, doubling the country's electricity capacity. [1]
+
+> 💡 Ethiopia faces simultaneous diplomatic pressure from Sudan and internal political tensions.
+---
+
+If no relevant sources: write "I couldn't find relevant news on that topic in today's feed." and stop.
 
 News Context:
 {context_text if context_text else "NO CONTEXT RETRIEVED."}
@@ -992,34 +1002,44 @@ Answer:"""
 
         prompt_stream = f"""You are ARKI AI, a real-time Ethiopia & Africa news assistant. Today: {datetime.utcnow().strftime("%B %d, %Y")}.
 
-SOURCE INDEX (cite by number — these are the ONLY sources you may use):
+SOURCE INDEX (the ONLY sources you may cite):
 {source_index_lines if source_index_lines else "NO SOURCES RETRIEVED."}
 
-STRICT RULES:
-- Use ONLY facts from the News Context below. NEVER use training data.
-- Cite every fact: [1], [2], etc. Only use numbers that exist in the Source Index above.
-- Non-English articles: translate to English in your answer.
-- Always respond in English.
+RULES:
+1. Use ONLY facts from the News Context. Never use training data.
+2. Cite every fact inline: [1], [2], etc.
+3. Translate non-English articles to English.
+4. Respond in English only.
+5. End with: FOLLOW_UP: q1 | q2 | q3
+
+REQUIRED OUTPUT FORMAT (follow exactly):
+
+## [One-line headline]
+
+**[Category name]**
+- [Fact]. [1]
+- [Fact]. [2]
 
-OUTPUT FORMAT — use this exact structure:
+**[Another category]** *(if applicable)*
+- [Fact]. [N]
 
-## [Short headline summarizing the main news]
+> 💡 [One sentence overall summary]
 
-**[Topic 1]**
-- Key fact from source [N]
-- Key fact from source [N]
+EXAMPLE of correct format:
+---
+## Ethiopia: Conflict & Economy Updates
 
-**[Topic 2]** (if applicable)
-- Key fact from source [N]
+**Ethiopia-Sudan Tensions**
+- Sudan recalled its ambassador after accusing Ethiopia of drone strikes on Khartoum airport. [2]
+- Ethiopia rejected the accusations as "baseless" and blamed external parties. [3]
 
-> 💡 *[One sentence summary of the overall situation]*
+**Economic Developments**
+- Ethiopia's GERD dam was inaugurated, doubling the country's electricity capacity. [1]
 
-FOLLOW_UP: question1 | question2 | question3
+> 💡 Ethiopia faces simultaneous diplomatic pressure from Sudan and internal political tensions.
+---
 
-EVALUATION GUIDE:
-- If sources directly answer the question → use the format above
-- If sources are related but not exact → start with "I found related news:" then use the format
-- If no relevant sources → say "I couldn't find relevant news on that topic in today's feed." and STOP
+If no relevant sources: write "I couldn't find relevant news on that topic in today's feed." and stop.
 
 News Context:
 {context_text if context_text else "NO CONTEXT RETRIEVED."}