Add host APIs for querying device ctx and remote heap pointer (#200)

* Add host APIs for querying device ctx and remote heap pointer

* Host API to query device pointer for ROCSHMEM_DEFAULT_CONTEXT,
  this is needed to support dynamic module initialization via device kernel
  library bitcode.
* Host API to query remote symmetric heap pointer that can be used in
  custom device kernel for RMA operations.

* Added rocshmem_ptr implementation within the Host Context class
* Enables pointer retrieval functionality for symmetric data objects
* Copy IPC pointers to host memory in RO host context

---------

Co-authored-by: avinashkethineedi <avinash.kethineedi@amd.com>

src/rocshmem_gpu.cpp

@@ -92,6 +92,21 @@ __device__ void rocshmem_query_thread(int *provided) {
 
 __device__ void rocshmem_wg_finalize() {}
 
+
+/******************************************************************************
+* These host APIs use Device side symbol - ROCSHMEM_CTX_DEFAULT so it needs 
+* to stay here to avoid getting pulled into other places in compilation
+******************************************************************************/
+
+__host__ void * rocshmem_get_device_ctx() {
+  void *ctx = nullptr;
+
+  CHECK_HIP(hipMemcpyFromSymbol(&ctx, HIP_SYMBOL(ROCSHMEM_CTX_DEFAULT),
+                             sizeof(rocshmem_ctx_t)));
+  return ctx;
+
+}
+
 /******************************************************************************
  ************************** Default Context Wrappers **************************
  *****************************************************************************/