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.

coreml/config.json

@@ -11,8 +11,6 @@
     {
       "file": "all_minilm_l6_v2_coreml_fp16.pte",
       "precision": "fp16",
-      "quantized": false,
-      "default": true,
       "methods": {
         "forward": {
           "inputs": [

vulkan/config.json

@@ -11,8 +11,6 @@
     {
       "file": "all_minilm_l6_v2_vulkan_fp16.pte",
       "precision": "fp16",
-      "quantized": false,
-      "default": true,
       "methods": {
         "forward": {
           "inputs": [

xnnpack/config.json

@@ -11,8 +11,6 @@
     {
       "file": "all_minilm_l6_v2_xnnpack_fp32.pte",
       "precision": "fp32",
-      "quantized": false,
-      "default": true,
       "methods": {
         "forward": {
           "inputs": [