calloc refs instead of malloc'ing them

Fix git_repository_message docs

This clarifies the docs for git_repository_message and also adds
to the tests to explicitly check NUL termination of data when the
output buffer is smaller than the message size.  There is a minor
behavior change so that a non-NULL output buffer will always be
NUL terminated (at length zero) if an error occurs.

Expose git_reference_dwim

Fix windows diff eofnl error

Config helpers for global/xdg config files

Extract this function out of the rev-parse code to be able to DWIM a
reference instead of its target.

The rules for which one to open is a bit silly, so let's make it
easier for our users.

When a repository is initialised, we need to probe to see if there is
a global config to load. If this is not the case, the user isn't able
to write to the global config without creating the backend and adding
it themselves, which is inconvenient and overly complex.

Unconditionally create and add a backend for the global config file
regardless of whether it exists as a convenience for users.

To enable this, we allow creating backends to files that do not exist
yet, changing the semantics somewhat, and making some tests invalid.

remote: correctly interpret tagopt '--tags'

When tagopt is set to '--tags', we should only take the default tags
refspec into account and ignore any configured ones.

Bring the code into compliance.

Diff patch bug fixes

When a patch contained an eofnl change (i.e. the last line either
gained or lost a newline), the oldno and newno line number values
for the lines in the last hunk of the patch were not useful.  This
makes them behave in a more expected manner.

This adds a new line origin constant for the special line that
is used when both files end without a newline.

In the course of writing the tests for this, I was having problems
with modifying a file but not having diff notice because it was
the same size and modified less than one second from the start of
the test, so I decided to start working on nanosecond timestamp
support.  This commit doesn't contain the nanosecond support, but
it contains the reorganization of maybe_modified and the hooks so
that if the nanosecond data were being read by stat() (or rather
being copied by git_index_entry__init_from_stat), then the nsec
would be taken into account.

This new stuff could probably use some more tests, although there
is some amount of it here.

fix git_branch_set_upstream for local branches

Fixed qsort_r() problem when targeting AmigaOS.

We fall back to the libgit2-provided insert sort as done for other
platforms.

Add git_commit_owner to the public interface

Just like git_tree_owner, etc.

Currently git_branch_set_upstream when passed a local branch
creates invalid configuration, for ex. if we setup branch
'tracking_master' to track local 'master' libgit2 generates
the following config

```
[branch "track_master"]
  remote = .
  merge = .refs/heads/track_master
```

The merge value is invalid and calling git_branch_upstream on
'tracking_master' results in invalid reference error.

It should do:

```
[branch "track_master"]
  remote = .
  merge = refs/heads/master
```

p_stat() should follow symlinks on windows

Make sure error messages are set for most error returns

By default do not set tagopt when cloning

allow checkout to proceed when a dir to be removed is in use (win32)

branch.h: fix typo in docs

Strict/lax commit API

Peeling to an invalid type is now checked via an assert so this
test is no longer relevant.