chore: version up to 2.80.40 and package with ADR-0008

package.json

@@ -2,7 +2,7 @@
   "name": "astra",
   "displayName": "Astra",
   "description": "The personal intelligence layer for Antigravity and VS Code. A private cognitive partner for deep project context, memory, and proactive strategic decision-making.",
-  "version": "2.80.39",
+  "version": "2.80.40",
   "publisher": "g1nation",
   "license": "MIT",
   "icon": "assets/icon.png",
@@ -225,9 +225,9 @@
         },
         "g1nation.smallModelContextCap": {
           "type": "number",
-          "default": 16384,
+          "default": 0,
           "minimum": 0,
-          "description": "When a genuinely tiny model (≤3B parameters, detected from the model name) is selected, budget the prompt against this smaller effective context window instead of g1nation.contextLength — very small models can emit an empty/EOS response on prompts that nominally fit but exceed their real capability. Does NOT apply to 4B+ models. Set 0 to disable entirely. Default: 16384"
+          "description": "Optional safety knob, OFF by default (0). Some very small models (≤3B) emit an empty/EOS response when given a prompt near their context window even though it nominally fits. If you observe that with a tiny model, set this to e.g. 8192–16384: for ≤3B models only, Astra then budgets the prompt against this smaller effective window instead of g1nation.contextLength. Never applies to 4B+ models. Leave 0 unless you actually hit the issue — it reduces the output-token budget. Default: 0 (disabled)"
         },
         "g1nation.autoContinueOnOutputLimit": {
           "type": "boolean",