Brunobkr/llama.cpp_AlgMor24_github

1```mermaid2%% MODEL Mode Data Flow (single model)3%% Detailed flows: ./flows/server-flow.mmd, ./flows/models-flow.mmd, ./flows/chat-flow.mmd4 5sequenceDiagram6    participant User as 👤 User7    participant UI as 🧩 UI8    participant Stores as 🗄️ Stores9    participant DB as 💾 IndexedDB10    participant API as 🌐 llama-server11 12    Note over User,API: 🚀 Initialization (see: server-flow.mmd, models-flow.mmd)13 14    UI->>Stores: initialize()15    Stores->>DB: load conversations16    Stores->>API: GET /props17    API-->>Stores: server config + modalities18    Stores->>API: GET /v1/models19    API-->>Stores: single model (auto-selected)20 21    Note over User,API: 💬 Chat Flow (see: chat-flow.mmd)22 23    User->>UI: send message24    UI->>Stores: sendMessage()25    Stores->>DB: save user message26    Stores->>API: POST /v1/chat/completions (stream)27    loop streaming28        API-->>Stores: SSE chunks29        Stores-->>UI: reactive update30    end31    API-->>Stores: done + timings32    Stores->>DB: save assistant message33 34    Note over User,API: 🔁 Regenerate35 36    User->>UI: regenerate37    Stores->>DB: create message branch38    Note right of Stores: same streaming flow39 40    Note over User,API: ⏹️ Stop41 42    User->>UI: stop43    Stores->>Stores: abort stream44    Stores->>DB: save partial response45```46