Clean up base constructor in model builder

.github/instructions/python-model-builder.instructions.md

@@ -11,4 +11,9 @@ When generating or reviewing code in the Python Model Builder (`src/python/py/mo
 
 Read both documents to understand the intended usage, supported models, design principles, and architectural constraints before suggesting or reviewing any code changes in this area.
 
-When a node is inserted into the model, prefer using `self.make_op_name` as the wrapper method for `self.make_node` + `self.make_value` calls.
+## Code Style Guidelines
+
+1. When a node is inserted into the model, prefer using `self.make_op_name` as the wrapper method for `self.make_node` + `self.make_value` calls.
+2. Ignore any CodeQL warnings about how an __init__ method calls an overridden method. These warnings are false positives and can be safely ignored. The warning message is: "this call to ABC in an initialization method is overwritten by XYZ".
+3. Find ways to reduce code duplication by reusing existing functionality and implementing common patterns.
+4. Discover ways to leverage the use of shared code in the base classes to avoid code duplication and improve maintainability.

src/python/py/models/builder.py

@@ -416,7 +416,7 @@ def get_args():
                 int4_block_size = 16/32/64/128/256: Specify the block size for int4 quantization (MatMulNBits).
                     Default value is 32.
                 qmoe_block_size = 16/32/64/128/256: Specify the block size for QMoE expert weights quantization.
-                    Default is 128 for trt-rtx, 32 for others. Supported EPs: cpu, webgpu, trt-rtx.
+                    Default is 128 for CUDA and TRT-RTX, 32 for others. Supported EPs: CPU, CUDA, WebGPU, TRT-RTX.
                 int4_is_symmetric = Quantize the weights symmetrically. Default is true.
                     If true, quantization is done to int4. If false, quantization is done to uint4.
                 int4_op_types_to_quantize = MatMul/Gather: Specify op types to target for int4 quantization.

src/python/py/models/builders/__init__.py

@@ -6,6 +6,11 @@
 # Modifications Copyright (C) 2026 Advanced Micro Devices, Inc. All rights reserved.
 # Portions of this file consist of AI generated content.
 # -------------------------------------------------------------------------
+import os.path
+import sys
+
+sys.path.append(os.path.dirname(__file__))
+
 from .base import Model
 from .chatglm import ChatGLMModel
 from .ernie import ErnieModel

src/python/py/models/builders/expansions/__init__.py

@@ -0,0 +1,10 @@
+# -------------------------------------------------------------------------
+# Copyright (c) Microsoft Corporation.  All rights reserved.
+# Licensed under the MIT License.  See License.txt in the project root for
+# license information.
+# -------------------------------------------------------------------------
+from .trt_rtx import TRT_RTX
+
+__all__ = [
+    "TRT_RTX",
+]