Keren/tensor slice insert alloc (#94)

This branch defines three new triton_gpu operations to partially solve #87. Below is an overview:

```
%tensor = triton_gpu.alloc_tensor : tensor<2x16x16xf16, #A>
%b = triton_gpu.insert_slice_async %a_ptr, %tensor, %offset {axis = 0 : i32, cache = 1 : i32, evict = 1 : i32, isVolatile = false} : tensor<16x16x!tt.ptr<f16>, #AL> -> tensor<2x16x16xf16, #A>
%c = triton_gpu.extract_slice %b, %offset {axis = 0 : i32} : tensor<2x16x16xf16, #A> -> tensor<16x16xf16, #A>
```

We plan to fully replace `copy_async` with `insert_slice_async`. **This hasn't been done yet.**

bin/triton-opt.cpp

@@ -35,7 +35,7 @@ int main(int argc, char **argv) {
   registry
       .insert<mlir::triton::TritonDialect, mlir::triton::gpu::TritonGPUDialect,
               mlir::arith::ArithmeticDialect, mlir::StandardOpsDialect,
-              mlir::scf::SCFDialect>();
+              mlir::scf::SCFDialect, mlir::gpu::GPUDialect>();
 
   return mlir::asMainReturnCode(mlir::MlirOptMain(
       argc, argv, "Triton (GPU) optimizer driver\n", registry));