[Backend] Vectorize Load/Store Ops (#86)

This PR does the following things:

- Code refactoring on Load and Store op codegen, rewrite with same logic
and share much code
- Support the vectorized load/store

python/triton/compiler.py

@@ -798,7 +798,8 @@ def optimize_tritongpu_ir(mod, num_stages):
     pm.add_tritongpu_pipeline_pass(num_stages)
     pm.add_canonicalizer_pass()
     pm.add_cse_pass()
-    # pm.add_triton_gpu_combine_pass()
+    pm.add_coalesce_pass()
+    pm.add_triton_gpu_combine_pass()
     pm.add_triton_gpu_verifier_pass()
     pm.run(mod)
     return mod