Add support for bool type in SplitToSequence (microsoft#24929)

### Description
Add support for `bool` type to address the issue below.

### Motivation and Context
This PR fixes microsoft#12286

Co-authored-by: Mauricio Cortazar <[email protected]>

Author: mauriciocm9

docs/OperatorKernels.md

@@ -437,7 +437,7 @@ Do not modify directly.*
 |||[13, 17]|**T** = tensor(bfloat16), tensor(bool), tensor(double), tensor(float), tensor(float16), tensor(int16), tensor(int32), tensor(int64), tensor(int8), tensor(string), tensor(uint16), tensor(uint32), tensor(uint64), tensor(uint8)|
 |||[11, 12]|**T** = tensor(bfloat16), tensor(bool), tensor(double), tensor(float), tensor(float16), tensor(int16), tensor(int32), tensor(int64), tensor(int8), tensor(string), tensor(uint16), tensor(uint32), tensor(uint64), tensor(uint8)|
 |||[2, 10]|**T** = tensor(bfloat16), tensor(bool), tensor(double), tensor(float), tensor(float16), tensor(int16), tensor(int32), tensor(int64), tensor(int8), tensor(string), tensor(uint16), tensor(uint32), tensor(uint64), tensor(uint8)|
-|SplitToSequence|*in* input:**T**<br> *in* split:**I**<br> *out* output_sequence:**S**|11+|**I** = tensor(int32), tensor(int64)<br/> **S** = seq(tensor(bfloat16)), seq(tensor(bool)), seq(tensor(double)), seq(tensor(float)), seq(tensor(float16)), seq(tensor(int16)), seq(tensor(int32)), seq(tensor(int64)), seq(tensor(int8)), seq(tensor(string)), seq(tensor(uint16)), seq(tensor(uint32)), seq(tensor(uint64)), seq(tensor(uint8))<br/> **T** = tensor(double), tensor(float), tensor(float16), tensor(int32), tensor(int64), tensor(string)|
+|SplitToSequence|*in* input:**T**<br> *in* split:**I**<br> *out* output_sequence:**S**|11+|**I** = tensor(int32), tensor(int64)<br/> **S** = seq(tensor(bfloat16)), seq(tensor(bool)), seq(tensor(double)), seq(tensor(float)), seq(tensor(float16)), seq(tensor(int16)), seq(tensor(int32)), seq(tensor(int64)), seq(tensor(int8)), seq(tensor(string)), seq(tensor(uint16)), seq(tensor(uint32)), seq(tensor(uint64)), seq(tensor(uint8))<br/> **T** = tensor(bool), tensor(double), tensor(float), tensor(float16), tensor(int32), tensor(int64), tensor(string)|
 |Sqrt|*in* X:**T**<br> *out* Y:**T**|13+|**T** = tensor(double), tensor(float)|
 |||[6, 12]|**T** = tensor(double), tensor(float)|
 |Squeeze|*in* data:**T**<br> *in* axes:**tensor(int64)**<br> *out* squeezed:**T**<br><br>or<br><br>*in* data:**T**<br> *out* squeezed:**T**|23+|**T** = tensor(bfloat16), tensor(bool), tensor(double), tensor(float), tensor(float16), tensor(int16), tensor(int32), tensor(int64), tensor(int8), tensor(string), tensor(uint16), tensor(uint32), tensor(uint64), tensor(uint8)|

onnxruntime/core/providers/cpu/sequence/sequence_ops.cc

@@ -339,7 +339,7 @@ ONNX_CPU_OPERATOR_KERNEL(
     11,
     KernelDefBuilder()
         .TypeConstraint("T",
-                        BuildKernelDefConstraints<float, MLFloat16, double, int32_t, int64_t, std::string>())
+                        BuildKernelDefConstraints<float, MLFloat16, double, int32_t, int64_t, bool, std::string>())
         .TypeConstraint("S", DataTypeImpl::AllSequenceTensorTypes())
         .TypeConstraint("I", BuildKernelDefConstraints<int32_t, int64_t>()),
     SplitToSequence);

onnxruntime/test/providers/cpu/sequence/sequence_ops_test.cc

@@ -536,5 +536,19 @@ TEST(SequenceOpsTest, SplitToSequence_PositiveAxisDontKeepDims) {
   test.AddSeqOutput("S2", output);
   test.Run();
 }
+
+TEST(SequenceOpsTest, SplitToSequence_BoolSplit) {
+  OpTester test("SplitToSequence", 11);
+  test.AddInput<bool>("input", {4, 2}, std::initializer_list<bool>({1, 1, 1, 1, 0, 0, 0, 0}));
+  int64_t axis = 0;
+  test.AddAttribute("axis", axis);
+  SeqTensors<bool> output;
+  output.AddTensor({1, 2}, {1, 1});
+  output.AddTensor({1, 2}, {1, 1});
+  output.AddTensor({1, 2}, {0, 0});
+  output.AddTensor({1, 2}, {0, 0});
+  test.AddSeqOutput("S2", output);
+  test.Run();
+}
 }  // namespace test
-}  // namespace onnxruntime
+}  // namespace onnxruntime