[Feature]: missing synchronisation functions

### Suggestion Description

The CUDA/HIP synchronisation with predicate functions are missing:
  - `int __syncthreads_count(int predicate);`
  - `int __syncthreads_and(int predicate);`
  - `int __syncthreads_or(int predicate);`

### Operating System

Any

### GPU

Any CPU

### ROCm Component

Kernel language