2.14.3-1

Add support for improved fault tolerance: non-blocking mode, new
init function with config, and ncclCommFinalize function.
Reintroduce collnet+chain algorithm, alongside collnet+direct.
Add LL protocol for intra-node P2P (on by default) and network
communication (off by default).
Use network instead of shared memory when performance is better.
Fix: wait for CUDA graph destroy before destroying comm with linked
graph resources.
Remove aggressive polling during enqueue.
Fix DMABUF fallback on MOFED 5.4 and earlier.

src/include/info.h

@@ -22,7 +22,8 @@ typedef enum : uint8_t {
   ncclPatternTreeUp,
   ncclPatternTreeDown,
   ncclPatternTreeUpDown,
-  ncclPatternCollTreeUpDown,
+  ncclPatternCollnetChain,
+  ncclPatternCollnetDirect,
   ncclPatternSend,
   ncclPatternRecv
 } ncclPattern_t;