Implement chat interface with message history and system prompt support; update frontend and tests accordingly

Co-authored-by: Copilot <copilot@github.com>

frontend/app/main.py

@@ -206,19 +206,64 @@ def generate():
 @app.route("/generate/text", methods=["GET", "POST"])
 @login_required
 def generate_text():
-    result = error = None
+    error = None
     token = session["access_token"]
+    chat_history: list[dict] = session.get("chat_history", [])
+    system_prompt: str = session.get("chat_system_prompt", "")
+    model: str = session.get("chat_model", "")
+
     if request.method == "POST":
-        resp = _api("POST", "/generate/text", token=token, json={
-            "model": request.form.get("model", "").strip(),
-            "prompt": request.form.get("prompt", "").strip(),
-        })
-        if resp.status_code == 200:
-            result = resp.json()
-        else:
-            error = resp.json().get("detail", "Generation failed.")
+        action = request.form.get("action", "send")
+
+        if action == "clear":
+            session.pop("chat_history", None)
+            session.pop("chat_system_prompt", None)
+            session.pop("chat_model", None)
+            return redirect(url_for("generate_text"))
+
+        prompt = request.form.get("prompt", "").strip()
+        model = request.form.get("model", "").strip()
+        system_prompt = request.form.get("system_prompt", "").strip()
+
+        # Persist model + system_prompt across turns
+        session["chat_model"] = model
+        session["chat_system_prompt"] = system_prompt
+
+        if prompt:
+            # Build messages: history (user/assistant only) + new user msg
+            messages = [m for m in chat_history if m["role"]
+                        in ("user", "assistant")]
+            messages.append({"role": "user", "content": prompt})
+
+            payload: dict = {
+                "model": model,
+                "messages": [{"role": m["role"], "content": m["content"]} for m in messages],
+            }
+            if system_prompt:
+                payload["system_prompt"] = system_prompt
+
+            resp = _api("POST", "/generate/text", token=token, json=payload)
+            if resp.status_code == 200:
+                data = resp.json()
+                chat_history = list(messages)
+                chat_history.append({"role": "assistant", "content": data["content"],
+                                     "usage": data.get("usage")})
+                session["chat_history"] = chat_history
+            else:
+                try:
+                    error = resp.json().get("detail", "Generation failed.")
+                except Exception:
+                    error = "Generation failed."
+
     models = _load_models(token, "text")
-    return render_template("generate_text.html", result=result, error=error, models=models)
+    return render_template(
+        "generate_text.html",
+        chat_history=session.get("chat_history", []),
+        error=error,
+        models=models,
+        system_prompt=system_prompt,
+        current_model=model,
+    )
 
 
 @app.route("/generate/image", methods=["GET", "POST"])