Swap frank-default-large to DFlash2 on Vulkan (PR #27342)

Replace the retired CIRU ROCmFPX runtime with upstream llama.cpp PR #27342
DFlash2 on the Vulkan backend:

- Target: unsloth Qwen3.8-27B-UD-Q8_K_XL.gguf (31.4 GB, already on disk)
- Drafter: z-lab Qwen3.8-27B-DFlash2-Q4_K_M.gguf (block size 8, top-k 16)
- Runtime: /home/dank/llama.cpp-dflash2/build-dflash2/bin/llama-server
  (GGML_VULKAN=ON; HIP lacks TOP_K so the stack runs on Vulkan0/RADV)
- spec: --spec-type draft-dflash --spec-draft-n-max 8 (clamps to 7)

Validated 2026-08-21 on scratch port 52626: mean acceptance 5.27,
health + reasoning completion OK.

Archive the retired CIRU .env/.serve script in .AGENTTRASH.
Update deployment.md, index.md, AGENTS.md; add docs/dflash2-large-vulkan.md.

scripts/serve-frank-large.sh

@@ -1,16 +1,15 @@
 #!/usr/bin/env bash
-# Serve jcbtc/Qwen3.8-27B-CIRU-ActiveFPX-PromptForge as frank-default-large
-# via the CIRU ROCmFPX llama.cpp runtime on Dankclaw (Strix Halo gfx1151).
+# Serve unsloth/Qwen3.8-27B-UD-Q8_K_XL as frank-default-large with a
+# z-lab DFlash2 drafter (Qwen3.8-27B-DFlash2-Q4_K_M) on the Vulkan backend.
 #
-# Runtime: the PromptForge llama-server built from ciru-ai/ROCmFPX tag
-# qwen3.8-activefpx-promptforge-v2.3 (commit 103af8b41) against the private
-# TheRock ROCm 7.15 SDK (kept in the retired ling3-ciru venv). The model is
-# a GGUF with optional PromptForge companion views (FFN/GDN .pfs) that enable
-# CIRU DualView prompt acceleration. Native depth-4 MTP speculative decoding.
+# Runtime: upstream llama.cpp PR #27342 (DFlash2 support) built with
+# GGML_VULKAN=ON for the Strix Halo gfx1151 (RADV). DFlash2 selects its
+# candidates through TOP_K + ARGMAX ops that the HIP backend does not
+# implement, so the whole stack runs on Vulkan0.
 #
-# The model card requires ROCm 7.15 (not the system ROCm 7.2.4) for the
-# advertised Strix Halo performance. LD_LIBRARY_PATH therefore points at the
-# 7.15 SDK libraries first. An older system ROCm must not precede them.
+# The target and the drafter share the Qwen3.8 arch. DFlash is lossless
+# under strict rejection sampling. --spec-draft-n-max clamps to the drafter's
+# trained block size (8).
 #
 # Tunables come from frank-default-large.env in this directory unless
 # FRANK_LARGE_ENV_FILE overrides that path.
@@ -26,26 +25,23 @@ if [[ -f "$ENV_FILE" ]]; then
 fi
 
 # --- defaults (used when the env file does not set them) -------------------
-: "${LLAMA_BIN:=/home/dank/ciru-qwen38/ROCmFPX/build-promptforge/bin/llama-server}"
-: "${MODEL_DIR:=/home/dank/models/CIRU-Qwen3.8-27B-ActiveFPX}"
-: "${MODEL_FILE:=Qwen3.8-27B-CIRU-ActiveFPX-PromptForge-v2.gguf}"
-: "${FFN_PFS:=Qwen3.8-27B-CIRU-ActiveFPX-PromptForge-FFN.pfs}"
-: "${GDN_PFS:=Qwen3.8-27B-CIRU-ActiveFPX-PromptForge-GDN.pfs}"
-: "${ROCM_DEVEL:=/home/dank/ling3-ciru/Ling-3.0-Flash-CIRU-int4-Strix-native/.venv/lib/python3.12/site-packages/_rocm_sdk_devel}"
-: "${ROCM_LIB:=/home/dank/ling3-ciru/Ling-3.0-Flash-CIRU-int4-Strix-native/.venv/lib/python3.12/site-packages/_rocm_sdk_libraries}"
-: "${ROCM_CORE:=/home/dank/ling3-ciru/Ling-3.0-Flash-CIRU-int4-Strix-native/.venv/lib/python3.12/site-packages/_rocm_sdk_core}"
+: "${LLAMA_BIN:=/home/dank/llama.cpp-dflash2/build-dflash2/bin/llama-server}"
+: "${MODEL_DIR:=/home/dank/models}"
+: "${MODEL_FILE:=Qwen3.8-27B-UD-Q8_K_XL.gguf}"
+: "${DRAFT_FILE:=Qwen3.8-27B-DFlash2-Q4_K_M.gguf}"
 : "${ALIAS:=frank-default-large}"
 : "${BIND_ADDR:=0.0.0.0}"
 : "${LISTEN_PORT:=52625}"
-: "${CTX_SIZE:=524288}"
-: "${PARALLEL:=2}"
+: "${CTX_SIZE:=262144}"
+: "${PARALLEL:=1}"
 : "${BATCH_SIZE:=4096}"
 : "${UBATCH_SIZE:=4096}"
 : "${CACHE_TYPE_K:=f16}"
 : "${CACHE_TYPE_V:=f16}"
 : "${CTX_CHECKPOINTS:=16}"
 : "${CACHE_RAM:=8192}"
-: "${SPEC_DRAFT_N_MAX:=4}"
+: "${SPEC_TYPE:=draft-dflash}"
+: "${SPEC_DRAFT_N_MAX:=8}"
 : "${SPEC_DRAFT_N_MIN:=0}"
 : "${SPEC_DRAFT_P_MIN:=0.0}"
 : "${SPEC_DRAFT_P_SPLIT:=0.10}"
@@ -56,22 +52,14 @@ fi
 : "${VERBOSITY:=4}"
 
 MODEL="$MODEL_DIR/$MODEL_FILE"
-export PROMPTFORGE_SIDECAR="$MODEL_DIR/$FFN_PFS"
-export PROMPTFORGE_GDN_SIDECAR="$MODEL_DIR/$GDN_PFS"
-export PROMPTFORGE_MODE="m2048_fused_tail1476"
-export GGML_CUDA_GRAPH_OPT=0
-export LLAMA_MTP_CPU_ARGMAX_FASTPATH=1
+DRAFT="$MODEL_DIR/$DRAFT_FILE"
 
 if [[ ! -f "$MODEL" ]]; then
     echo "missing required file: $MODEL" >&2
     exit 1
 fi
-if [[ ! -f "$PROMPTFORGE_SIDECAR" ]]; then
-    echo "missing PromptForge FFN companion: $PROMPTFORGE_SIDECAR" >&2
-    exit 1
-fi
-if [[ ! -f "$PROMPTFORGE_GDN_SIDECAR" ]]; then
-    echo "missing PromptForge GDN companion: $PROMPTFORGE_GDN_SIDECAR" >&2
+if [[ ! -f "$DRAFT" ]]; then
+    echo "missing DFlash2 drafter: $DRAFT" >&2
     exit 1
 fi
 if [[ ! -x "$LLAMA_BIN" ]]; then
@@ -79,16 +67,12 @@ if [[ ! -x "$LLAMA_BIN" ]]; then
     exit 1
 fi
 
-# ROCm 7.15 SDK libraries must precede any system ROCm in LD_LIBRARY_PATH.
-export ROCM_PATH="$ROCM_DEVEL"
-export LD_LIBRARY_PATH="${ROCM_LIB}/lib:${ROCM_CORE}/lib:${ROCM_DEVEL}/lib${LD_LIBRARY_PATH:+:$LD_LIBRARY_PATH}"
-
 exec "$LLAMA_BIN" \
     --model "$MODEL" \
     --alias "$ALIAS" \
     --host "$BIND_ADDR" \
     --port "$LISTEN_PORT" \
-    -dev ROCm0 \
+    -dev Vulkan0 \
     -ngl 999 \
     -c "$CTX_SIZE" \
     -b "$BATCH_SIZE" \
@@ -102,8 +86,9 @@ exec "$LLAMA_BIN" \
     --cache-prompt \
     --metrics \
     --no-mmap \
-    --spec-type draft-mtp \
-    --spec-draft-device ROCm0 \
+    --spec-type "$SPEC_TYPE" \
+    --spec-draft-model "$DRAFT" \
+    --spec-draft-device Vulkan0 \
     --spec-draft-ngl 999 \
     --spec-draft-type-k f16 \
     --spec-draft-type-v f16 \
@@ -111,10 +96,9 @@ exec "$LLAMA_BIN" \
     --spec-draft-n-min "$SPEC_DRAFT_N_MIN" \
     --spec-draft-p-min "$SPEC_DRAFT_P_MIN" \
     --spec-draft-p-split "$SPEC_DRAFT_P_SPLIT" \
-    --spec-draft-backend-sampling \
     --temp "$TEMPERATURE" \
     --top-p "$TOP_P" \
     --reasoning "$REASONING" \
     --reasoning-format "$REASONING_FORMAT" \
     --reasoning-budget -1 \
-    -lv "$VERBOSITY"
+    -lv "$VERBOSITY"