distributed collectives (#1519)

* feat: world

* feat: tests

* feat: no more backwards

* feat: recv into

* feat: whoops

* feat: test in ci

* feat: some debug logging

* feat: workflow naming

* feat: need to set pythonpath

* feat: just send to same device

* feat: allreduce

* feat: test

* feat: need contiguous

* feat: test in ci

* feat: exit with correct code

* feat: don't need that

* feat: opencl wait_for just doesn't work

* feat: synchronize on out

* feat: try?

* feat: try again?

* feat: add extra realizes

* feat: print

* feat: seed

* feat: tol

* feat: test ones and zeros

* feat: remove print

* feat: are you just flaky

* feat: seperate scatter and gather?

* feat: just try synchronizing

* feat: remove print again

* feat: bring back difference

* feat: no sync

* feat: revert that

* feat: back to wait_for

* fix: typo

tinygrad/runtime/ops_gpu.py

@@ -48,7 +48,7 @@ class CLBuffer(RawBufferCopyInOut):
     assert not self.dtype.name.startswith("image"), f"can't copyout images {self.dtype}"
     buf = cl.Buffer(CL.cl_ctxs[self._buf.device], cl.mem_flags.WRITE_ONLY | cl.mem_flags.USE_HOST_PTR, 0, hostbuf=x.data)
     mapped, event = cl.enqueue_map_buffer(CL.cl_queue[self._buf.device], buf, cl.map_flags.WRITE, 0, self.size, dtype=self.dtype.np, is_blocking=False)
-    with mapped.base: cl.enqueue_copy(CL.cl_queue[self._buf.device], mapped, self._buf, is_blocking=True, wait_for=[event])
+    with mapped.base: cl.enqueue_copy(CL.cl_queue[self._buf.device], mapped, self._buf, is_blocking=True, wait_for=[event] + ([self.event] if hasattr(self, "event") else []))
 
 class CLProgram:
   def __init__(self, name:str, prg:str, binary=False, argdtypes=None, options=None):