Support group API with only SEND or RECV operations

[pkourdis]

Summary

Group API of oneCCL expects the collectives grouped to finish before group_end() so the following pattern is not supported:

group_start();
send(send_buf_ptr, sendcount, ...);
group_end();
group_start();
recv(recv_buf_ptr, sendcount, ...);
group_end();

However, the above pattern is supported by NCCL.

Reproducer

See internal MLSL-3958.

Affected projects

XCCL backend for TorchComms.