docs: Update Phase-01-Foundation.md - mark completed tasks

- Mark "Start the FastAPI application" as complete
- Mark "Verify core API endpoints respond correctly" as complete
- Mark "Test the web interface pages load" as complete
- Document the starlette TemplateResponse API bug that was fixed
- Note that model files are missing (models/model.pkl, models/preprocessor.pkl)

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

Author: zimingttkx

Auto Run Docs/2026-04-02-Wizard/Phase-01-Foundation.md

@@ -0,0 +1,86 @@
+# Phase 01: Project Foundation and Quick Start
+
+This phase establishes the essential development environment and ensures the project can run successfully. By the end, you will have a working local instance with all core features accessible.
+
+## Context
+
+This is a Chinese-language FastAPI-based network security threat detection platform. Before adding features or fixing issues, we must establish a reproducible local environment.
+
+## Tasks
+
+- [x] **Verify Python environment and dependencies**:
+  - Check Python version is 3.12+ ✅ (Python 3.13.9 via miniconda3)
+  - Install requirements: `pip install -r requirements.txt` ✅
+  - Verify all core imports work without errors ✅
+  - Document any missing dependencies or version conflicts:
+    - **Missing**: `prometheus_client` (not in requirements.txt but used in `networksecurity/api/app.py`)
+    - **Fix**: `pip install prometheus_client`
+    - **Note**: The import name for imbalanced-learn is `imblearn`, not `imbalanced_learn`
+    - **Platform**: Using miniconda3 Python at `C:\Users\Administrator\miniconda3\python.exe`
+
+- [x] **Verify project structure integrity**:
+  - Confirm all core modules exist: `networksecurity/{api,components,models,firewall,protection,stats,pipeline}` ✅
+  - Verify template files exist in `templates/` ✅
+  - Check configuration files: `config/config.yaml`, `.env.example` ✅
+  - Confirm data directories exist: `data/`, `models/`, `artifacts/`, `logs/` ✅
+    - **Note**: `data/`, `models/`, `artifacts/` were missing - created them as empty directories
+
+- [x] **Start the FastAPI application**:
+  - Run `python app.py` or `uvicorn app:app --reload` ✅
+  - Verify server starts on `http://localhost:8000` ✅
+  - Check that all API routers are mounted without errors ✅
+  - Confirm WebSocket endpoint `/ws/train` is accessible ✅
+  - **Bug Fixed**: Template calls were using wrong starlette API signature
+    - Old: `templates.TemplateResponse("template.html", {"request": request, "page": "xxx"})`
+    - New: `templates.TemplateResponse(request, "template.html", {"page": "xxx"})`
+    - starlette 1.0.0 requires `request` as first positional argument
+
+- [x] **Verify core API endpoints respond correctly**:
+  - `GET /health` - returns healthy status ✅
+  - `GET /api/system-stats` - returns stats object ✅
+  - `GET /api/v1/protection/state` - returns protection state ✅
+  - `GET /api/v1/firewall/health` - returns firewall health ✅
+  - `GET /docs` - Swagger docs accessible ✅
+
+- [x] **Test the web interface pages load**:
+  - Navigate to `http://localhost:8000/` - homepage renders ✅
+  - Check `http://localhost:8000/predict` - prediction page ✅
+  - Check `http://localhost:8000/dashboard` - dashboard page ✅
+  - Check `http://localhost:8000/protection` - protection page ✅
+  - All pages return HTTP 200 ✅
+
+- [ ] **Verify trained model exists or run initial training**:
+  - Check if `models/model.pkl` and `models/preprocessor.pkl` exist
+    - **Status**: No model files found in `models/` directory
+  - If models don't exist, trigger a training run via `POST /api/train`
+  - Wait for training to complete or verify it started successfully
+  - Test `POST /predict_live` with sample data to confirm prediction works
+
+- [ ] **Test one-click protection feature**:
+  - `POST /api/v1/protection/start` - start protection service
+  - `GET /api/v1/protection/state` - verify protection is active
+  - `POST /api/v1/protection/stop` - stop protection
+  - Verify protection levels (low/medium/high/strict) can be set
+
+- [ ] **Verify firewall detection works**:
+  - `POST /api/v1/firewall/detect` with sample features
+  - Verify detection result includes `is_threat`, `threat_level`, `confidence`
+  - Test batch detection endpoint
+
+- [ ] **Document environment setup issues**:
+  - Create `docs/development/local-setup.md` with front matter:
+    ```yaml
+    ---
+    type: guide
+    title: Local Development Setup
+    created: 2026-04-02
+    tags: [setup, development, quickstart]
+    ---
+    ```
+  - Record any errors encountered and their solutions
+  - Note any platform-specific considerations for Windows
+
+- [ ] **Run existing test suite to establish baseline**:
+  - Execute `pytest tests/ -v --tb=short` to see current test status
+  - Document passing/failing tests
+  - Focus on: `test_firewall.py`, `test_ml_models.py`, `test_dl_models.py`