[ExecuTorch][WebGPU] Add fused SDPA (sdpa_with_kv_cache) with dynamic input_pos

Pull Request resolved: #20086

Adds the fused `sdpa_with_kv_cache` op (QK attention-weights, softmax, attention-output sub-kernels over the KV cache), composing the three enablers below it: the base graph's inter-dispatch buffer passing (scratch buffers + multi-pass execute), the `update_cache` op, and the SymInt live-scalar mechanism. The QK/softmax/AV kernels mirror the Vulkan reference's flat-index/GQA/causal-mask math (NCHW, buffer-only, fp32).

`input_pos` is consumed dynamically via the SymInt mechanism: the op reads `symint_buffer()` as a uniform, sizes its scratch + dispatches for the max context length, and registers a resize hook so a single delegate runs an autoregressive decode loop (feed only the new token + advancing `input_pos`) instead of a fixed baked position. Mirrors the Vulkan SymInt = live uniform-buffer design.

Tests live in the stacked test-suite diff above (clean op diff here).

Authored with assistance from Claude.
ghstack-source-id: 392609088
@exported-using-ghexport

Differential Revision: [D107595125](https://our.internmc.facebook.com/intern/diff/D107595125/)

Author: JulianCloudNTH

backends/webgpu/CMakeLists.txt

@@ -35,6 +35,7 @@ set(WEBGPU_SRCS
     runtime/ops/add/BinaryOp.cpp
     runtime/ops/rms_norm/RmsNorm.cpp
     runtime/ops/update_cache/UpdateCache.cpp
+    runtime/ops/sdpa/Sdpa.cpp
     runtime/ops/select_as_symint/SelectAsSymint.cpp
 )
 

backends/webgpu/runtime/WebGPUGraph.h

@@ -106,6 +106,9 @@ class WebGPUGraph {
   int64_t get_int(int id) const {
     return ints_[id];
   }
+  bool get_bool(int id) const {
+    return bools_[id];
+  }
 
   // Live-scalar (SymInt) API; mirrors the Vulkan SymInt/ParamsBuffer UBO.
   // set_symint writes the buffer + marks dirty only if the value changed.