[MLIR] Split ExecutionEngine Initialization out of ctor into an explicit method call (#153524)

Retry landing https://github.com/llvm/llvm-project/pull/153373
## Major changes from previous attempt
- remove the test in CAPI because no existing tests in CAPI deal with
sanitizer exemptions
- update `mlir/docs/Dialects/GPU.md` to reflect the new behavior: load
GPU binary in global ctors, instead of loading them at call site.
- skip the test on Aarch64 since we have an issue with initialization there

---------

Co-authored-by: Mehdi Amini <joker.eph@gmail.com>

mlir/python/mlir/_mlir_libs/_mlirExecutionEngine.pyi

@@ -19,5 +19,6 @@ class ExecutionEngine:
     def dump_to_object_file(self, file_name: str) -> None: ...
     def raw_lookup(self, func_name: str) -> int: ...
     def raw_register_runtime(self, name: str, callback: object) -> None: ...
+    def init() -> None: ...
     @property
     def _CAPIPtr(self) -> object: ...