Restructure into a monorepo: models, VS Code extension, daemon, shared

- models/: GGUFs (Git LFS), manifest.json, model card, Apache-2.0 LICENSE + NOTICE
- extensions/vscode/: the Local Autocomplete extension (v0.5.0, publisher harvmaster)
- daemon/: acd completion daemon (bundled into the extension; also runs standalone)
- shared/: prompt layout + symbol extraction shared with the training pipeline
- scripts/: make-model-release.mjs, publish-extension-release.mjs (Gitea releases)
- README with quick start, settings, architecture, build and release instructions;
  MIT license for code; CHANGELOG

The default model manifest URL moves to models/manifest.json.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

daemon/acd.config.json

@@ -0,0 +1,62 @@
+{
+  "host": "127.0.0.1",
+  "port": 7071,
+  "tokenizer": {
+    "json": "assets/qwen-tokenizer.json",
+    "config": "assets/qwen-tokenizer_config.json"
+  },
+  "llamaServer": {
+    "binary": "llama-server",
+    "port": 7072,
+    "ctx": 8192,
+    "args": [
+      "-ngl",
+      "99"
+    ]
+  },
+  "models": {
+    "stage-c": {
+      "path": "../models/stage-c-q8_0.gguf",
+      "order": "spm",
+      "modes": [
+        "symbols",
+        "simple"
+      ],
+      "defaultMode": "simple"
+    },
+    "stage-b": {
+      "path": "../models/stage-b-q8_0.gguf",
+      "order": "spm",
+      "modes": [
+        "raw"
+      ],
+      "defaultMode": "raw"
+    }
+  },
+  "defaultModel": "stage-c",
+  "prompt": {
+    "prefixTokens": 1536,
+    "suffixTokens": 512,
+    "prefixChunkLines": 32,
+    "maxLines": 8,
+    "nPredictSingleLine": 48,
+    "nPredictMultiLine": 128
+  },
+  "symbols": {
+    "budget": 2048,
+    "waitMs": 150,
+    "members": false
+  },
+  "thisRepair": {
+    "enabled": true,
+    "waitMs": 100
+  },
+  "telemetry": {
+    "enabled": false,
+    "dir": "telemetry"
+  },
+  "recording": {
+    "dir": "recordings"
+  },
+  "stateFile": "acd.state.json"
+}