Various extensions to the vectorizer

Includes additions to the AstVectorizer:

• Vectorize assignments to vector symbols
• Vectorize loops nested within the vectorization axis if all bounds are independent of the axis
• Vectorize array subscripts if all indices and the array expression are independent of the axis
• Handle empty nodes (e.g. comments)

Additions to the x86 instruction set:

• Intrinsics for basic math functions (log, abs, etc.)
• Type conversions

And to the SelectIntrinsics pass:

• Add option to implemenet type conversions using __builtin_convertvector which is supported by most compilers instead of inserting intrinsics. Useful because not all conversions are supported natively on all CPUs.

src/pystencils/backend/platforms/x86.py

@@ -159,6 +159,9 @@ class X86VectorCpu(GenericVectorCpu):
         self, expr: PsExpression, operands: Sequence[PsExpression]
     ) -> PsExpression:
         match expr:
+            case PsCast(target_type, _) if target_type == operands[0].get_dtype():
+                assert len(operands) == 1
+                return operands[0]
             case PsUnOp() | PsBinOp():
                 func = _x86_op_intrin(self._vector_arch, expr, expr.get_dtype())
                 intrinsic = func(*operands)
@@ -262,7 +265,9 @@ class X86VectorCpu(GenericVectorCpu):
             addr: PsExpression = PsAddressOf(PsMemAcc(acc.pointer, acc.offset))
             if addr_type:
                 addr = PsCast(addr_type, addr)
-            return load_func(addr)
+            intrinsic = load_func(addr)
+            intrinsic.dtype = load_func.return_type
+            return intrinsic
         else:
             raise NotImplementedError("Gather loads not implemented yet.")
 
@@ -274,7 +279,9 @@ class X86VectorCpu(GenericVectorCpu):
             addr: PsExpression = PsAddressOf(PsMemAcc(acc.pointer, acc.offset))
             if addr_type:
                 addr = PsCast(addr_type, addr)
-            return store_func(addr, arg)
+            intrinsic = store_func(addr, arg)
+            intrinsic.dtype = store_func.return_type
+            return intrinsic
         else:
             raise NotImplementedError("Scatter stores not implemented yet.")