Stop restating quantized, and drop the unused default

quantized duplicated precisions.json, which is authoritative for the
quantized/non-quantized partition. default expressed a per-repository
recommendation nothing consumed: the library resolves its own per-platform
default. Both are gone from the schema.

0_8b/xnnpack/config.json

@@ -14,8 +14,6 @@
     {
       "file": "qwen_3_5_0_8b_xnnpack_8da4w.pte",
       "precision": "8da4w",
-      "quantized": true,
-      "default": true,
       "methods": {
         "forward": {
           "inputs": [

2b/xnnpack/config.json

@@ -14,8 +14,6 @@
     {
       "file": "qwen_3_5_2b_xnnpack_8da4w.pte",
       "precision": "8da4w",
-      "quantized": true,
-      "default": true,
       "methods": {
         "forward": {
           "inputs": [

vl_0_8b/xnnpack/config.json

@@ -15,8 +15,6 @@
     {
       "file": "qwen_3_5_vl_0_8b_xnnpack_8da4w.pte",
       "precision": "8da4w",
-      "quantized": true,
-      "default": true,
       "methods": {
         "forward": {
           "inputs": [

vl_2b/xnnpack/config.json

@@ -15,8 +15,6 @@
     {
       "file": "qwen_3_5_vl_2b_xnnpack_8da4w.pte",
       "precision": "8da4w",
-      "quantized": true,
-      "default": true,
       "methods": {
         "forward": {
           "inputs": [