Use SIMD operations in InitBlkUnroll/CopyBlkUnroll and increase unroll limit up to 128 bytes (#61030)

* Use SIMD operations in CodeGen::genCodeForInitBlkUnroll()

* Use SIMD operations in CodeGen::genCodeForCpBlkUnroll()

* Remove restrictions on offset values in Lowering::ContainBlockStoreAddress()

* Remove unused macros in src/coreclr/jit/instr.h

Author: echesakov