Update api_ref_quantization docs (#1619)

Author: andrewor14

docs/source/api_ref_quantization.rst

@@ -6,24 +6,43 @@ torchao.quantization
 
 .. currentmodule:: torchao.quantization
 
+Main Quantization APIs
+----------------------
+
 .. autosummary::
     :toctree: generated/
     :nosignatures:
 
-    autoquant    
     quantize_
-    int8_dynamic_activation_int4_weight
-    int8_dynamic_activation_int8_weight
+    autoquant    
+
+Quantization APIs for quantize_
+-------------------------------
+
+.. autosummary::
+    :toctree: generated/
+    :nosignatures:
+
     int4_weight_only
     int8_weight_only
+    int8_dynamic_activation_int4_weight
+    int8_dynamic_activation_int8_weight
+    uintx_weight_only
+    gemlite_uintx_weight_only
+    intx_quantization_aware_training
+    from_intx_quantization_aware_training
     float8_weight_only
     float8_dynamic_activation_float8_weight
     float8_static_activation_float8_weight
-    uintx_weight_only
     fpx_weight_only
-    to_linear_activation_quantized
-    swap_linear_with_smooth_fq_linear
-    smooth_fq_linear_to_inference
+
+Quantization Primitives
+-----------------------
+
+.. autosummary::
+    :toctree: generated/
+    :nosignatures:
+
     choose_qparams_affine
     choose_qparams_affine_with_min_max
     choose_qparams_affine_floatx
@@ -40,3 +59,16 @@ torchao.quantization
     ZeroPointDomain
     TorchAODType
 
+..
+  TODO: delete these?
+
+Other
+-----
+
+.. autosummary::
+    :toctree: generated/
+    :nosignatures:
+
+    to_linear_activation_quantized
+    swap_linear_with_smooth_fq_linear
+    smooth_fq_linear_to_inference

docs/source/api_ref_sparsity.rst

@@ -10,9 +10,9 @@ torchao.sparsity
     :toctree: generated/
     :nosignatures:
 
-    WandaSparsifier
-    PerChannelNormObserver
-    apply_fake_sparsity
     sparsify_
     semi_sparse_weight
     int8_dynamic_activation_int8_semi_sparse_weight
+    apply_fake_sparsity
+    WandaSparsifier
+    PerChannelNormObserver