chore: update llama-cpp to b7898 and opencode to v1.1.48

- Update llama-cpp from b7867 to b7898
- Update opencode from v1.1.12 to v1.1.48 with improved build process:
  - Replace custom bundle script with official script/build.ts
  - Add shell completion support
  - Add version check testing
  - Simplify node_modules handling
- Update llama-swap service config with new llama.cpp options
- Clarify opencode agent testing workflow in developer and reviewer configs

modules/nixos/services/llama-swap/config.nix

@@ -19,8 +19,8 @@ in
           --port ''${PORT} \
           -m /mnt/ssd/Models/GLM/GLM-4.7-Flash-UD-Q4_K_XL.gguf \
           -c ''${ctx} \
+          -ctk q8_0 \
           --jinja \
-          --threads -1 \
           --temp 0.7 \
           --top-p 1.0 \
           --min-p 0.01 \
@@ -34,6 +34,11 @@ in
       env = [ "GGML_CUDA_ENABLE_UNIFIED_MEMORY=1" ];
     };
 
+    # --spec-type ngram-mod \
+    # --spec-ngram-size-n 24 \
+    # --draft-min 48 \
+    # --draft-max 64 \
+
     # https://huggingface.co/unsloth/Devstral-Small-2-24B-Instruct-2512-GGUF/tree/main
     "devstral-small-2-instruct" = {
       name = "Devstral Small 2 (24B) - Instruct";