handle error in TCKEYREQ during TCINDXREQ better

  Dont be too aggressive in Dbtc::nodeFailCheckTransaction
  let it timeout by 1, so that it does not assert that it has waited too long
  old impl. set timeotu value to 0, making timeout = (ctcTimer - 0) which could be quite big.

  add testprg for SR case aswell

  add new testpgrom for bug#18612 to autotest

  remove bug#18385 from autotest as it only works on 2 node clusters

  Fix timeout during ABORT when ZABORT_TIMEOUT_BREAK is outstanding

  Partial system restart, can not try to start with higher GCI that own
  even if knowing about a higher number

  Use variable waitfor_response_timeout (depending on TransactionDeadLockTimeout)
  When getting 4012, set NeedAbort and ReleaseOnClose

  dont't allow node to join cluster until all nodes has completed failure handling

  Error in abort handling in TC when timeout during abort
  

  Fix typo in autotest script

  Fix error in NF during NR

  add testcase to autotest

  incorrect handling of writeTuple with multi op transaction

  add more tests to verify correctness of backups

  Redo abort handling according to descr. in Backup.txt
bug#9960 - ndb backup
      increase wait completed timeout to 48 hours

  increased timeout as it is exceeded every now and then giving erroneous indication on error

  crash if scan is poped from queue before all attrinfo has arrived

    optimized for fewer timeouts
    changed for consistency in "time"
tests
    moved SR_UNDO tests to basic
    added testBackup -n BackupBank

testTransaction -n *Scan*
testTimeout -n *