Merge pull request !1073 from yao_yf/reshape_bug_fix

Merge pull request !1054 from yangzhenzhang/add-slice-shape-for-param-info

Merge pull request !1051 from yao_yf/auto_parallel_reshape_reconstruct

Merge pull request !1028 from Chong/memctl

Merge pull request !1023 from lichen/add_gatherv2_distributed_op

Merge pull request !897 from Chong/avril

Merge pull request !727 from Chong/cost

Merge pull request !728 from yao_yf/strategy_checkpoint_extend

Merge pull request !746 from lirongzhen1/bp_reducescatter

Merge pull request !494 from zhoufeng/cmake-sub

Merge pull request !645 from yao_yf/auto_parallel_prelu_support_broadcast

Merge pull request !643 from lichen/support_reshape_parameter

Merge pull request !586 from lirongzhen1/allreduce_fusion

Merge pull request !264 from xychow/remove-unnecessary-cache-and-add-cache

Merge pull request !620 from Xiaoda/fix-codex-bot-warnings

Merge pull request !556 from Chong/partition

add cache for PythonPrimEvaluator. Be careful that
 the infer function of PythonPrimitive in python code should
 be idempotent.

Merge pull request !405 from Chong/str-gen

Merge pull request !394 from Chong/parser

Merge pull request !87 from gziyan/dev_alltoall

ALLGATHER, and REDUCESCATTER with different factors; change the BETA and
GAMMA value in cost model.

clean clang format errors and cpplint errors

add some test cases for histogram summary operator

Merge pull request !385 from Xiaoda/modify-communicaiton-cost-of-operators-and-redistribution