This branch introduces a new transaction API.  The key features are:
  - top-level functions in transaction -- get(), commit(), abort()
  - explicit transaction manager objects
  - Transaction objects are used for exactly one transaction
  - support for transaction synchronizers

The changes here are still provisional, but we want to get them off an
obscure branch and onto the head for further development.

Bug reported by Dieter Maurer.

The change required several of the read-conflict tests to be updated
to specify that they don't want MVCC support.  It was basically
accidental that they passed before this change.  Note that we don't
have any tests of code that gets a read conflict in MVCC mode, which
would only be possible after a pack() or in some other weird case.

- Renamed updateCodeTimestamp() to resetCaches().

- Added a test of resetCaches().

- Removed a spurious comment about chaining connections together.

The DB's invalidate() method takes a set of oids corresponding to all
the changes from a data manager for one transaction.  All the objects
are invalidated at once.

Add a few tests in testZODB of the new code.  The tests just cover
corner cases, because I can't think of a sensible way to test the
atomicity.  When it has failed in the past, it's been caused by
nearly-impossible to reproduce data races.

This fix needs to be backported to Zope 2.6, but only after assessing
how significant an impact the API change will have.

This change adds a new method, setLocalTransaction(), to the
Connection class.  ZODB applications can call this method to bind
transactions to connections rather than threads.  This is especially
useful for GUI applications, which often have only one thread but
multiple independent activities within that thread (generally one per
window).  Thanks to Christian Reis for championing this feature.

Applications that take advantage of this feature should not use the
get_transaction() function.  Until now, ZODB itself sometimes assumed
get_transaction() was the only way to get the transaction.  Minor
corrections have been added.  The ZODB test suite, on the other hand,
can continue to use get_transaction(), since it is free to assume that
transactions are bound to threads.

updated, which resulted in one whole cache being kept for no reason.

Almost entirely concerned with adding ZPL blocks.

XXX Not sure if berkeley still works.

# NB: commit() is responsible for calling tpc_begin() on the storage.
# It uses self._begun to track whether it has been called.  When
# self._begun is None, it has not been called.

# This arrangement allows us to handle the special case of a
# transaction with no modified objects.  It is possible for
# registration to be occur unintentionally and for a persistent
# object to compensate by making itself as unchanged.  When this
# happens, it's possible to commit a transaction with no modified
# objects.

# Since tpc_begin() may raise a ReadOnlyError, don't call it if there
# are no objects.  This avoids spurious (?) errors when working with
# a read-only storage.

Add code to handle this in Connection's tpc_begin() and commit()
methods.

Add two tests in testZODB.

StorageTestBase.removefs() will attempt to remove files with all the
possible extensions that FileStorage will create.  It will raise
os.error for any error except ENOENT.

Remove many variants of removefs() implemented in the various test
suites.

so makes the full suite brittle.  Cleaned out all the places where unit
tests change sys.path.  (Use the PYTHONPATH environment variable instead.)

Also brought SearchIndex tests up to date, which I didn't really need to do
since SearchIndex is deprecated, but I did it as part of figuring out the
failed Interface tests, so I might as well check in my work. ;-)  One of the
test modules in SearchIndex forgets to define a global named "Dummy",
actually, so all the tests were failing before this checkin anyway.