The old implementation had two issues:

1. OIDs that were too short as to be ambiguous were not being handled
properly.

2. If the last OID to expand in the array was missing from the ODB, we
would leak a `GIT_ENOTFOUND` error code from the function.

Introduce `git_odb_expand_ids`

Take (and write to) an array of a struct, `git_odb_expand_id`.

Support for ssh+git and git+ssh protocols

commit: split creating the commit and writing it out

Sometimes you want to create a commit but not write it out to the
objectdb immediately. For these cases, provide a new function to
retrieve the buffer instead of having to go through the db.

Enable nanosecond resolution by default

Instead of hoping that we can get a racy entry by going real fast
and praying real hard, just create a racy entry.

Update link to Pro Git's Git internals chapter

If the underlying filesystem doesn't support better than one
second resolution, then don't expect that turning on `GIT_USE_NSEC`
does anything magical to change that.

Query the object database for multiple objects at a time, given their
object ID (which may be abbreviated) and optional type.

When looking up an abbreviated oid, show the actual (abbreviated) oid
the caller passed instead of a full (but ambiguously truncated) oid.

Update unit test to use newfangled `st_ctime_nsec`, which provides
indirection to the platform-correct name.

treebuilder: don't try to verify submodules exist in the odb

Submodules don't exist in the objectdb and the code is making us try to
look for a blob with its commit id, which is obviously not going to
work.

Skip the test if the user wants to insert a submodule.

ssh: initialize libssh2

We should have been doing this, but it initializes itself upon first
use, which works as long as nobody's doing concurrent network
operations. Initialize it on our init to make sure it's not getting
initialized concurrently.

Expand OpenSSL and libssh2 thread safety documentation

Nanosecond resolution is now the default in git itself.  Enable this
as our default as well.

test: make sure we retry the auth callback on all platforms

xdiff: fix memleak on error case

If the caller has provided bad authentication, give them another
apportunity to get it right until they give up. This brings WinHTTP in
line with the other transports.

We were missing this test on Windows, which meant we didn't notice that
we never fixed the single authentication attempt it tries, nor its wrong
return code.

Enable this for the unix platforms as well over HTTP. We previously were
doing it locally but disabled it on OS X due to issues with its sshd not
accepting password authentication.

Stricter object dependency checking during creation

Commit 3d1abc5a fixes a memory leak in the xdiff code. In the
process of upstreaming the fix it was pointed out by Johannes
Schindelin that there is another memory leak present (see [1]).

Fix the second memory leak by applying the upstream fix to our
code base.

[1]: http://thread.gmane.org/gmane.comp.version-control.git/287034

Few p_getaddrinfo fixes

Use legitimate (existing) object IDs in tests so that we have the
ability to turn on strict object validation when running tests.

This allows lighter weight validation in `git_object__is_valid` that
does not require reading the entire object.

When `GIT_OPT_ENABLE_STRICT_OBJECT_CREATION` is turned on, validate
the index entries given to `git_index_add`.

travis ci: enable debug pool for valgrind builds