Fix variable number of inputs in mps_example.py (#918)

Summary:
- fix models with variable number of inputs (>1). Needed for the models `add`, `linear`, `add_mul` where the inputs array is usually made out of 2/3 tensors.
- llama2 is also working correctly through MPS delegate. Set the `atol` to 1e-4 when doing the bundled program comparison check with the eager output from CPU.

Pull Request resolved: #918

Reviewed By: kirklandsign

Differential Revision: D50301445

Pulled By: kimishpatel

fbshipit-source-id: 38576363de60fc45dabc92833723cf8f293cc7c5

Author: DenisVieriu97

examples/apple/mps/executor_runner/mps_executor_runner.mm

@@ -481,10 +481,11 @@ MemoryManager memory_manager(
         strstr(model_path, "emformer_transcribe")  ||
         strstr(model_path, "emformer_join")        ||
         strstr(model_path, "edsr")                 ||
+        strstr(model_path, "llama2")               ||
         strstr(model_path, "ic3")                  ||
         strstr(model_path, "ic4")) {
         atol = 1e-04;
-      }
+    }
     status = torch::executor::util::VerifyResultWithBundledExpectedOutput(
         *method,
         file_data->data(),

examples/apple/mps/scripts/mps_example.py

@@ -75,8 +75,8 @@ def __init__(self):
             super().__init__()
             self.mps_module = lowered_module
 
-        def forward(self, input_args):
-            return self.mps_module(input_args)
+        def forward(self, *input_args):
+            return self.mps_module(*input_args)
 
     executorch_program = (
         exir.capture(