[GENERAL] Merged v1.0alpha into master. Added features are:

- A100 support via mma.16816
- Thread swizzling for conflict-free shared memory accesses without
padding
- Complete overhaul of the LLVM code generation in
codegen/selection/generator.cc to remove overengineering
- Added debugging capabilities in the Python binding
- Compilation error for kernels that spill

cmake/FindLLVM.cmake

@@ -28,7 +28,8 @@
 # We also want an user-specified LLVM_ROOT_DIR to take precedence over the
 # system default locations such as /usr/local/bin. Executing find_program()
 # multiples times is the approach recommended in the docs.
-set(llvm_config_names llvm-config-10 llvm-config-10.0 llvm-config100
+set(llvm_config_names llvm-config-11 llvm-config-11.0
+                      llvm-config-10 llvm-config-10.0 llvm-config100
                       llvm-config-9 llvm-config-9.0 llvm-config90
                       llvm-config-8 llvm-config-8.0 llvm-config80
                       llvm-config)