clients, but ZEO clients haven't provided very good protection,
leading to cache corruption.  We'll hopefully fix these client issues,
which cause other problems beside cache corruption, but it seems
prudent to provide low-level cache protection.

was astronomical.  Also greatly simplified the code.

refactoring to simplify the data structures to fix a serious memory
bug: the cache uses waaaay the heck too much.

- Cache records are written in several stems.  If a process exits
    after writing begins and before it is finishes, the cache will be
    corrupt on restart.  The way records are writted was changed to
    make cache record updates atomic.

- There was no lock file to prevent opening a cache multiple times
    at once, which would lead to corruption.  Persistent caches now
    use lock files, in the same way that file storages do.

- A bug in the cache-opening logic led to cache failure in the
    unlikely event that a cache has no free blocks.

When using ZEO Client Storages, Errors occured when trying to store
objects too big to fit in the ZEO cache file.

setUp():  Use a much smaller (than default) ZEO client cache.

The current default is 200MB, and at least testSerialization reads
the entire cache file into one giant string.  That grossly boosts
the process highwater mark when running the tests.  Windows reclaims
the memory after, but other platforms may not.

Cool:  testSerialization was one of the slowest tests before, but
is 100x faster now.

More cleanup.  Notable:

ClientCache:  Removed the .tid attribute.  It was never updated; the
contained FileCache instance actually keeps track of the last tid.
This had the happy side effect of fixing bugs in testSerialization:
it was accessing .tid directly, and so was _always_ comparing None to
None.  Changing the test to use .getLastTid() instead means it's now
testing what it always intended to test.

FileCache:  Removed some unused private attributes for tracking
internal statistics.

More code and comment cleanups.  Notable changes:

FileCache:  removed the currentsize attribute.  It was initialized
to 0 but never updated.  Some tests referenced it, but since it was
always 0 those tests weren't getting any good from it.  Don't
recall (or never knew) what its intended purpose may have been.

FileCache.remove():  Repaired major bug.  This mistakenly stored
the "this disk block is free now" status byte into the start of
the serialized Object record's end_tid field (incorrect seek
offset).  The in-memory structures were correct then, but got out
of synch with the disk file; the latter then still claimed to have
info for a "live" object revision, but that revision was actually
dead, and the info on disk had a corrupt end_tid value.