Add Linux development station role with systemd services for code-server and OpenCode

- Implemented handlers for restarting code-server and OpenCode services.
- Created main tasks for setting up a Linux development station, including user management, package installation, and configuration.
- Added validation for OpenCode server credentials and code-server password.
- Introduced tool package management for npm, uv, and go installers.
- Developed templates for systemd service units for code-server and OpenCode.
- Included scripts for starting code-server and OpenCode with SSH agent support.
- Updated inventory and Terraform configuration to include new development host.

ansible/host_vars/gpu-01.lab.alexpires.me.yml

@@ -118,25 +118,25 @@ llama_server_devices:
   - "/dev/dri:/dev/dri:rw"
 
 llama_server_models:
-  # TODO: replace revision values with pinned commit SHAs for strict reproducibility.
+  # # TODO: replace revision values with pinned commit SHAs for strict reproducibility.
   - name: "gemma-4-12b-q8-0"
     model_id: "unsloth/gemma-4-12b-it-GGUF"
     quant: "Q8_0"
     revision: "main"
-    preset:
-      ctx-size: 98304
+    # preset:
+    #   ctx-size: 98304
   - name: "gemma-4-26b-a4b-ud-iq4-xs"
     model_id: "unsloth/gemma-4-26B-A4B-it-GGUF"
     quant: "UD-IQ4_XS"
     revision: "main"
-    preset:
-      ctx-size: 98304
+    # preset:
+    #   ctx-size: 98304
   - name: "gemma-4-31b-iq4-xs"
     model_id: "unsloth/gemma-4-31B-it-GGUF"
     quant: "IQ4_XS"
     revision: "main"
-    preset:
-      ctx-size: 98304
+    # preset:
+    #   ctx-size: 98304
   - name: "gpt-oss-20b-q4-0"
     model_id: "unsloth/gpt-oss-20b-GGUF"
     quant: "Q4_0"
@@ -149,44 +149,56 @@ llama_server_models:
     model_id: "unsloth/Qwen3.5-27B-GGUF"
     quant: "Q4_K_M"
     revision: "main"
-    preset:
-      ctx-size: 98304
+    # preset:
+    #   ctx-size: 98304
   - name: "qwen3-5-35b-a3b-q4-k-m"
     model_id: "unsloth/Qwen3.5-35B-A3B-GGUF"
     quant: "Q4_K_M"
     revision: "main"
-    preset:
-      ctx-size: 98304
+    # preset:
+    #   ctx-size: 98304
   - name: "qwen3-5-9b-q8-0"
     model_id: "unsloth/Qwen3.5-9B-GGUF"
     quant: "Q8_0"
     revision: "main"
-    preset:
-      ctx-size: 98304
+    # preset:
+    #   ctx-size: 98304
   - name: "qwen3-6-27b-q4-k-m"
     model_id: "unsloth/Qwen3.6-27B-GGUF"
     quant: "Q4_K_M"
     revision: "main"
-    preset:
-      ctx-size: 98304
+    # preset:
+    #   ctx-size: 98304
+  - name: "qwen3-6-35b-a3b-ud-q4-k-m"
+    model_id: "unsloth/Qwen3.6-35B-A3B-GGUF"
+    quant: "UD-Q4_K_M"
+    revision: "main"
+    # preset:
+    #   ctx-size: 98304
   - name: "qwen3-6-27b-mtp-q4-0"
     model_id: "unsloth/Qwen3.6-27B-MTP-GGUF"
     quant: "Q4_0"
     revision: "main"
     preset:
-      ctx-size: 98304
+      temperature: 0.6
+      top_p: 0.95
+      top_k: 20
+      min_p: 0.0
+      presence_penalty: 0.0
+      repeat_penalty: 1.0
+      # ctx-size: 98304
   - name: "qwen3-6-35b-a3b-mtp-ud-q4-m"
     model_id: "unsloth/Qwen3.6-35B-A3B-MTP-GGUF"
     quant: "UD-Q4_K_M"
     revision: "main"
     preset:
-      ctx-size: 98304
-  - name: "qwen3-6-35b-a3b-ud-q4-k-m"
-    model_id: "unsloth/Qwen3.6-35B-A3B-GGUF"
-    quant: "UD-Q4_K_M"
-    revision: "main"
-    preset:
-      ctx-size: 98304
+      temperature: 0.6
+      top_p: 0.95
+      top_k: 20
+      min_p: 0.0
+      presence_penalty: 0.0
+      repeat_penalty: 1.0
+    #   ctx-size: 98304
 
 llama_server_argv_extra:
   [
@@ -195,15 +207,15 @@ llama_server_argv_extra:
     "-np",
     -1,
     "-b",
-    1024,
+    2048,
     "-ub",
     512,
     "-fa",
     "on",
     "-ctk",
-    "q4_0",
-    "-ctv",
     "q8_0",
+    "-ctv",
+    "q4_0",
     "-cram",
     -1,
     "-sm",
@@ -216,11 +228,14 @@ llama_server_argv_extra:
     "off",
     "-mg",
     1,
+    "--timeout",
+    1800,
     "--mmap",
     "--spec-type",
     "draft-mtp",
     "--spec-draft-n-max",
     2,
+    "--swa-full",
   ]
 
 auto_suspend_enabled: true