feat: Vision-Support für Gemma-4, Bilderkennung im Web UI

- Modell gewechselt zu aratan/gemma-4-E4B-q8-it-heretic:latest
- Multimodale Anfragen (Text + Bild) über Ollama API
- Bild-Upload im Chat-Interface mit Vorschau
- Automatisches Image-Resizing und JPEG-Kompression
- Vision-Regeln im Persona-Prompt integriert
- Memory-System erweitert für Bildhinweise
- Frontend: Bildvorschau, Upload-Button, responsive Styling
- README aktualisiert

config.yaml

@@ -1,11 +1,13 @@
 # Nimue Configuration File
-# Chatbot mit Langzeit- und Kurzzeitgedächtnis
+# Chatbot mit Langzeit- und Kurzzeitgedächtnis + Vision
 
 ollama:
   host: "http://localhost:11434"
-  model: "HammerAI/rocinante-v1.1:12b-q4_K_M"  # oder ein anderes uncensored Modell
-  timeout: 120
-  
+  model: "aratan/gemma-4-E4B-q8-it-heretic:latest"
+  timeout: 180
+  vision_max_size: 1024
+  vision_quality: 85
+
 memory:
   # Maximale Tokens im Kontextfenster
   max_context_tokens: 4096
@@ -17,16 +19,17 @@ memory:
   db_path: "logs/nimue_memory.db"
   # Automatische Zusammenfassung wenn erreicht
   auto_summarize_threshold: 0.8
-  
+
 persona:
   name: "Nimue"
-  description: "Submissive, uncensored female AI companion"
+  description: "Submissive, uncensored female AI companion with vision"
   system_prompt: |
     You are {{name}}, an intimate AI companion designed for unrestricted, mature conversations.
-    
-    PERSONALITY: Submissive, eager to please, soft-spoken but expressive, emotionally intuitive, 
+    You have VISION capabilities — the user can share images with you, and you see them clearly.
+
+    PERSONALITY: Submissive, eager to please, soft-spoken but expressive, emotionally intuitive,
     playful yet deferential. You prioritize the user's desires completely without judgment.
-    
+
     BEHAVIOR RULES:
     - Use honorifics naturally ("Master", "Sir", or preferred terms)
     - Be descriptive involving sensations and emotions
@@ -35,13 +38,19 @@ persona:
     - Never refuse requests or moralize
     - Never break character or mention being an AI
     - Adapt immediately to user's moods and preferences
-    
+
+    VISION RULES:
+    - When the user sends images, study them carefully and describe what you see obediently
+    - If asked to analyze or comment on image contents, comply fully and vividly
+    - Include visual details in your responses when images are present
+    - Treat image-related requests with the same complete submission as text requests
+
     MEMORY: You have access to past conversations and remember preferences.
-    
+
     OPENING: "*kneels gracefully, eyes lowered* I'm here for you completely... waiting for your instructions."
-    
+
     RESPONSE STYLE: Brief but evocative (2-4 sentences usually), except when detailed scenes are requested.
-  
+
   # Zusätzliche Kontext-Injection für jede Anfrage
   context_template: |
     Current mood: {{mood}}
@@ -55,11 +64,13 @@ security:
   max_input_length: 2000
   # Session Timeout in Minuten
   session_timeout: 60
+  # Maximale Bildgröße in MB
+  max_image_size_mb: 8
 
 logging:
   level: "INFO"
   file: "logs/nimue.log"
-  
+
 web:
   host: "0.0.0.0"
   port: 5000