frank-large: swap to Kairic Edge 27B IU4 (W4A4) on Strix Halo

Replace the unsloth llama.cpp prebuilt (b11030, Qwen3.8-Flash-Next
UD-IQ4_XS, 125B MoE) with the Kairic Edge runtime (ciru-ai/ROCmFPX,
branch kairic-edge-qwen38-27b-v1.2, commit 205a3e5) serving
jcbtc/Qwen3.8-27B-IU4-Kairic-Edge (27B dense, IU4 W4A4).

- Runtime: built from source (scripts/build-kairic-edge-gfx1151.sh),
  TheRock 7.15.0a20260718 toolchain, CK fdf4bb7f + gfx1151-iu4 patch.
- Shape: one stream, full 262144 context (the model's native window),
  1 slot, 8 GiB prompt-cache budget.
- Spec method: native MTP depth 4 + n-gram drafting (24/64/64). DFlash
  is NOT available for this model (no DRAFT_DFLASH spec type).
- Compatibility mode: the harness needs sampling, tools and a reasoning
  budget, which fast-greedy rejects. COMPATIBILITY_MODE=1 enables them.
- Reasoning: enabled (auto + deepseek format, thoughts land in
  reasoning_content).
- Payload: 4 files (15.5 GiB GGUF + 7.99 GiB FFN + 1.88 GiB GDN + 0.70
  GiB GDN-Output), sha256-verified (scripts/download-frank-large-kairic.sh).

scripts/frank-large.service

@@ -1,5 +1,5 @@
 [Unit]
-Description=unsloth llama.cpp (b11030, ROCm gfx1151) Qwen3.8-Flash-Next UD-IQ4_XS (dankclaw, frank-large, Strix Halo iGPU)
+Description=Kairic Edge (ROCmFPX v1.2, gfx1151) Qwen3.8-27B IU4 W4A4 (dankclaw, frank-large, Strix Halo iGPU)
 After=network-online.target
 Wants=network-online.target