Convert scalar_to_vector to ISLE (AArch64) (#4401)

* Convert `scalar_to_vector` to ISLE (AArch64)

Converted the exisiting implementation of `scalar_to_vector` for AArch64 to
ISLE.

Copyright (c) 2022 Arm Limited

* Add support for floats and fix FpuExtend

- Added rules to cover `f32 -> f32x4` and `f64 -> f64x2` for
`scalar_to_vector`
- Added tests for `scalar_to_vector` on floats.
- Corrected an invalid instruction emitted by `FpuExtend` on 64-bit
values.

Copyright (c) 2022 Arm Limited

cranelift/codegen/src/isa/aarch64/lower.isle

@@ -121,6 +121,20 @@
 (rule (lower (has_type $I128 (iconcat lo hi)))
       (output (value_regs lo hi)))
 
+;;;; Rules for `scalar_to_vector` ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+
+(rule (lower (has_type $F32X4 (scalar_to_vector x)))
+      (fpu_extend x (ScalarSize.Size32)))
+
+(rule (lower (has_type $F64X2 (scalar_to_vector x)))
+      (fpu_extend x (ScalarSize.Size64)))
+
+(rule (lower (scalar_to_vector x @ (value_type (ty_int_bool_64 _))))
+      (mov_to_fpu x (ScalarSize.Size64)))
+
+(rule (lower (scalar_to_vector x @ (value_type (int_bool_fits_in_32 _))))
+      (mov_to_fpu (put_in_reg_zext32 x) (ScalarSize.Size32)))
+
 ;;;; Rules for `iadd_pairwise` ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 
 (rule (lower (has_type $I16X8 (iadd_pairwise (swiden_low x) (swiden_high y))))