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.

mlx/config.json

@@ -17,8 +17,6 @@
         "vocoder": "vocoder_mlx_fp32.pte"
       },
       "precision": "fp32",
-      "quantized": false,
-      "default": true,
       "methods": {
         "duration_predictor": {
           "forward": {

vulkan/config.json

@@ -17,8 +17,6 @@
         "vocoder": "vocoder_vulkan_fp16.pte"
       },
       "precision": "fp16",
-      "quantized": false,
-      "default": true,
       "methods": {
         "duration_predictor": {
           "forward": {

xnnpack/config.json

@@ -17,8 +17,6 @@
         "vocoder": "vocoder_xnnpack_fp32.pte"
       },
       "precision": "fp32",
-      "quantized": false,
-      "default": true,
       "methods": {
         "duration_predictor": {
           "forward": {