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/gemma4_test_result.txt

@@ -0,0 +1,8 @@
+
+==================================================
+ Gemma4 26B Q4_K_M  5-Run Results:
+   AVG:  74.65 t/s
+   BEST: 75.07 t/s
+   MIN:  74.27 t/s
+   Runs: ['74.59', '74.68', '74.65', '75.07', '74.27']
+==================================================