NEO did not ensure that all data and metadata are written on disk before
tpc_finish, and it was for example vulnerable to ENOSPC errors.
In other words, some work had to be moved to tpc_vote:

- In tpc_vote, all involved storage nodes are now asked to write all metadata
  to ttrans/tobj and _commit_. Because the final tid is not known yet, the tid
  column of ttrans and tobj now contains NULL and the ttid respectively.

- In tpc_finish, AskLockInformation is still required for read locking,
  ttrans.tid is updated with the final value and this change is _committed_.

- The verification phase is greatly simplified, more reliable and faster. For
  all voted transactions, we can know if a tpc_finish was started by getting
  the final tid from the ttid, either from ttrans or from trans. And we know
  that such transactions can't be partial so we don't need to check oids.

So in addition to minimizing the risk of failures during tpc_finish, we also
fix a bug causing the verification phase to discard transactions with objects
for which readCurrent was called.

On performance side:

- Although tpc_vote now asks all involved storages, instead of only those
  storing the transaction metadata, the client has been improved to do this
  in parallel. The additional commits are also all done in parallel.

- A possible improvement to compensate the additional commits is to delay the
  commit done by the unlock.

- By minimizing the time to lock transactions, objects are read-locked for a
  much shorter period. This is even more important that locked transactions
  must be unlocked in the same order.

Transactions with too many modified objects will now timeout inside tpc_vote
instead of tpc_finish. Of course, such transactions may still cause other
transaction to timeout in tpc_finish.

We can never receive several answers from the same node.

testVerification is dropped for the same reason as for testEvent and most of
testConnection, since there is much incoming changes for verification.

Such kind of test has never helped to detect regressions and any bug in
EpollEventManager would be quickly reported by other tests.

testConnection may go the same way if it keeps annoying me too much.

Before it was only possible to specify them in a configuration file.

They're decompressed automatically to temporary files.

That seems too verbose now but it was difficult to know the ports of nodes
when only warnings were output.

It was broken since review of logging
(commit 1fce5cc4).

Export:
- Remove leftover warning about a bug that was fixed in
  commit e76af297
- In neomigrate script, open NEO storage read-only.
- IStorageIteration is already implemented.

Import:
- Review comments.
- In neomigrate script, warn that IStorageRestoreable is not implemented.
- Do not call 'close' method on source iterator. BaseStorage does not do it and
  this is not part of ZODB API. In the case of FileStorage, resource are freed
  automatically during garbage collection.

In particular, a cluster may now starts automatically so 'simple' can't know
whether the cluster is in an expected state or not.

It would be unreliable to resume printing logs using a condition on date,
so the format of log files is changed to add primary keys that number records.

Since lib.protocol has been rewritten to be semantic-oriented and
duplicates a lot less code, this test is just bloated (testing the same
code over and over) and a maintenance burden.

The main goal of this patch is to keep all DEBUG logs and packet logger enabled
without exploding disk usage.
This is done by keeping the last 16 MB (by default) of debugging information in
a RAM buffer, and to emit it to an SQLite DB upon RTMIN signal or in case of
warning and more severe log.

Implementation is also cleaned up for better integration within a framework
or if run standalone. NEO logger is now a direct child of root handler.
Only warnings and more severe logs are forwarded to root handler.

A new script 'neolog' is added to pretty-print the contents of the SQLite log.

In unit tests, logging events are not buffered but emitted immediately.
When a test passes, payloads of all exchanged packets are discarded to reduce
disk usage on test bots.

This slows down performance tests by about 15 % because even if nothing is
written to disk, debug and packet log records are now always rendered.

Replication is also fully reimplemented:
- It is not done anymore on whole partitions.
- It runs at lowest priority not to degrades performance for client nodes.

Schema of MySQL table is changed to optimize storage layout: rows are now
grouped by age, for good partial replication performance.
This certainly also speeds up simple loads/stores.

Commit 03ae3309 broke unit tests.

This reverts check that unneeded options are passed, and only skips run()
when used from 'neostorage' script.

Do not actually start storage node when told to reset database content.
Refuse to reset database content when non-required parameters are provided.