register sdpa operator with model config

[lanluo-nvidia]

Description

Fixes # (issue)

Type of change

Please delete options that are not relevant and/or add your own.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

Checklist:

• My code follows the style guidelines of this project (You can use the linters)
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas and hacks
• I have made corresponding changes to the documentation
• I have added tests to verify my fix or my feature
• New and existing unit tests pass locally with my changes
• I have added the relevant labels to my PR in so that relevant reviewers are notified

[github-actions]

There are some changes that do not conform to Python style guidelines:

--- /home/runner/work/TensorRT/TensorRT/tools/llm/torchtrt_ext/register_sdpa.py	2025-08-25 20:35:29.149375+00:00
+++ /home/runner/work/TensorRT/TensorRT/tools/llm/torchtrt_ext/register_sdpa.py	2025-08-25 20:35:59.784651+00:00
@@ -162,6 +162,6 @@

    gm = clean_up_graph_after_modifications(gm)
    new_output_tensors = create_random_output_tensors(new_outputs)
    new_out_spec = pytree.tree_flatten(new_output_tensors)[1]
    gm._out_spec = new_out_spec
-    return gm
\ No newline at end of file
+    return gm
--- /home/runner/work/TensorRT/TensorRT/tools/llm/torchtrt_ext/sdpa_converter.py	2025-08-25 20:35:29.149375+00:00
+++ /home/runner/work/TensorRT/TensorRT/tools/llm/torchtrt_ext/sdpa_converter.py	2025-08-25 20:35:59.940021+00:00
@@ -286,11 +286,10 @@
                    false_input = mm
                    output_layer = if_layer.add_output(
                        true_input.get_output(0), false_input.get_output(0)
                    )
                    scaled_add_attn_bias = output_layer.get_output(0)
-    

    softmax = impl.normalization.softmax(
        ctx, target, source_ir, name + "_softmax", scaled_add_attn_bias, -1, False
    )
    if use_fp32_acc: