feat: Variet Engine v1.0 + 5-model tuning complete

Phase 01 (LLM Tuning):
- Gemma4 26B: 74.65 t/s (fast)
- Qwen 35B: 61.62 t/s (balanced)
- Gemma4 31B: 16.0 t/s (deep-coder)
- Qwen 27B: 16.7 t/s (deep-logic)
- Qwen 122B: 8.95 t/s (ultra, GPU 1 only)

Phase 02 (API Engine):
- FastAPI reverse proxy on port 8000
- /engine/switch hot-swap with 503 protection
- config/engine_models.json as single source of truth
- Replaced 4 individual .bat files with unified engine

File cleanup:
- scripts/ 85 files -> 9 + _archive/
- Root .bat files -> _archive/

scripts/_archive/results/quality_result_qwen35b.json

@@ -0,0 +1,124 @@
+[
+  {
+    "id": "code_generate",
+    "category": "coding",
+    "name": "Python 함수 생성",
+    "model": "qwen35b",
+    "response": "",
+    "tokens": 800,
+    "time": 12.7,
+    "tps": 62.97,
+    "eval_criteria": [
+      "correctness",
+      "type_hints",
+      "docstring",
+      "edge_cases"
+    ]
+  },
+  {
+    "id": "code_debug",
+    "category": "coding",
+    "name": "버그 찾기 & 수정",
+    "model": "qwen35b",
+    "response": "Here's the corrected version of the function that returns a dictionary of duplicate values with their occurrence counts:\n\n```python\ndef find_duplicates(arr):\n    count = {}\n    for item in arr:\n        count[item] = count.get(item, 0) + 1\n    \n    return {item: cnt for item, cnt in count.items() if cnt > 1}\n```\n\n### Explanation of Fixes:\n\n1. **Correct Output Format:** The function now returns a dictionary where each key is a duplicate value and the corresponding value is its total count of occurrences.\n2. **Handles Empty Lists:** The function works correctly with an empty list, returning an empty dictionary `{}`.\n3. **Accurate Duplicate Detection:** Only values that appear more than once are included in the result.\n\n### Example Usage:\n\n```python\nprint(find_duplicates([1, 2, 2, 3, 3, 3]))  # Output: {2: 2, 3: 3}\nprint(find_duplicates([]))                 # Output: {}\nprint(find_duplicates([1, 1, 1]))          # Output: {1: 3}\nprint(find_duplicates([1, 2, 3]))          # Output: {}\n```\n\nThis implementation is efficient, readable, and meets the specified requirements.",
+    "tokens": 788,
+    "time": 12.8,
+    "tps": 61.58,
+    "eval_criteria": [
+      "bug_identified",
+      "correct_fix",
+      "clean_code"
+    ]
+  },
+  {
+    "id": "code_refactor",
+    "category": "coding",
+    "name": "TypeScript 리팩토링",
+    "model": "qwen35b",
+    "response": "",
+    "tokens": 800,
+    "time": 13.05,
+    "tps": 61.3,
+    "eval_criteria": [
+      "types",
+      "error_handling",
+      "backoff",
+      "production_quality"
+    ]
+  },
+  {
+    "id": "korean_schedule",
+    "category": "assistant_kr",
+    "name": "한국어 일정 관리",
+    "model": "qwen35b",
+    "response": "",
+    "tokens": 800,
+    "time": 12.89,
+    "tps": 62.08,
+    "eval_criteria": [
+      "korean_fluency",
+      "schedule_analysis",
+      "practical_advice"
+    ]
+  },
+  {
+    "id": "korean_email",
+    "category": "assistant_kr",
+    "name": "한국어 이메일 요약",
+    "model": "qwen35b",
+    "response": "",
+    "tokens": 800,
+    "time": 13.06,
+    "tps": 61.27,
+    "eval_criteria": [
+      "korean_summary",
+      "action_items",
+      "conciseness"
+    ]
+  },
+  {
+    "id": "tool_calling",
+    "category": "tool_use",
+    "name": "Function Calling (JSON)",
+    "model": "qwen35b",
+    "response": "",
+    "tokens": 800,
+    "time": 12.92,
+    "tps": 61.9,
+    "eval_criteria": [
+      "correct_sequence",
+      "valid_json",
+      "complete_args"
+    ]
+  },
+  {
+    "id": "structured_output",
+    "category": "tool_use",
+    "name": "구조화 출력 (JSON)",
+    "model": "qwen35b",
+    "response": "",
+    "tokens": 800,
+    "time": 13.04,
+    "tps": 61.34,
+    "eval_criteria": [
+      "correct_parsing",
+      "valid_json",
+      "completeness"
+    ]
+  },
+  {
+    "id": "reasoning",
+    "category": "reasoning",
+    "name": "논리 추론",
+    "model": "qwen35b",
+    "response": "",
+    "tokens": 800,
+    "time": 12.86,
+    "tps": 62.21,
+    "eval_criteria": [
+      "correct_answer",
+      "clear_steps",
+      "math_accuracy"
+    ]
+  }
+]