This was just wrong.  Added a test that verifying patch line
numbers even for hunks further into a file and then fixed the
algorithm.  I needed to add a little extra state into the patch
so that I could track old and new file numbers independently,
but it should be okay.

Introduce a delta base cache

Some regex warning fixes

It turns out the indexers have been ignoring the pack's free function
and leaking data. Plug that.

Currently limited to 16MB (like git) and to objects up to 1MB in
size.

refspec: prevent git_refspec__free() from segfaulting

Many delta bases are re-used. Cache them to avoid inflating the same
data repeatedly.

This version doesn't limit the amount of entries to store, so it can
end up using a considerable amound of memory.

Fix libgit2/libgit2sharp#247

There are different solutions to the problem. In this change, we
define an UNUSED macro that maps to __attribute__((unused)) when
compiling with gcc. Otherwise it is a NOOP. We apply this macro
in all function headers for each parameter value that is not used
within the function body.

The change is local to regex.

Add GIT_CHECKOUT_DISABLE_PATHSPEC_MATCH

This adds an option to checkout a la the diff option to turn off
fnmatch evaluation for pathspec entries.  This can be useful to
make sure your "pattern" in really interpretted as an exact file
match only.

Expose stock user/pass credential utility

REUC needs to handle empty sides

Sane refresh logic for #1180

Libgit2 for AmigaOS4

All the ODB backends have a specific refresh interface. When reading an
object, first we attempt every single backend: if the read fails, then
we refresh all the backends and retry the read one more time to see if
the object has appeared.

Don't call pthread_exit() in the callback of the fetch example

Compilers that are not aware that pthread_exit() does not return
issue a warning when compiling the present code. This change
exchanges the call to pthread_exit() with a simple return
statement. According to the pthread specification this is
equivalent.

Before it was compiled when CMake was actually run on AmigaOS.

push: properly handle tags

Currently, push doesn't really handle tags when queueing objects. Fix
it.

Resolve crash with diff against empty file

It is not legal inside our `p_mmap` function to mmap a zero length
file.  This adds a test that exercises that case inside diff and
fixes the code path where we would try to do that.

The fix turns out not to be a lot of code since our default file
content is already initialized to "" which works in this case.

Fixes #1210