While we are confident about the size of an int in architectures we're
likely to care about, the index format is defined by the exact size of
the fields. Use the definitions which show the exact width of the entry
fields.

As part of that, bring back 32-bit time and size fields, which currently
are 64 bits wide and can bring a false sense of security in how much
data they really store. Document that these fields are not to be taken
as authoritative.

git_index_add_frombuffer enables now to store a memory buffer in the odb
and to store an entry in the index directly if the index is attached to a
repository.

This makes them show up in the reference, even if the text itself isn't
the most descriptive.

These have been found with

    grep -Przon '\n\ntypedef struct.*?\{' -- include
    grep -Przon '\n\ntypedef enum.*?\{' -- include

The documentation has shown this as a single enum for a long time. These
should in fact be two enums. One with the bits for the flags and another
with the bits for the extended flags.

These are some issues I found while playing around with the new parser
for docurium.

This reorganized the diff OID calculation to make it easier to
correctly update the stat cache during a diff once the flags to
do so are enabled.

This includes marking the path of a git_index_entry as const so
we can make a "fake" git_index_entry with a "const char *" path
and not get warnings.  I was a little surprised at how unobtrusive
this change was, but I think it's probably a good thing.

I introduced a leak into conflict cleanup by removing items from
inside the git_vector_remove_matching call.  This simplifies the
code to just use one common way for the two conflict cleanup APIs.

When an index has an active snapshot, removing an item can cause
an error (inserting into the deferred deletion vector), so I made
the git_index_conflict_cleanup API return an error code.  I felt
like this wasn't so bad since it is just like the other APIs.

I fixed up a couple of comments while I was changing the header.

Again, laying groundwork for some index iterator changes, this
contains a bunch of code refactorings for index internals that
should make it easier down the line to add locking around index
modifications.  Also this removes the redundant prefix_position
function and fixes some potential memory leaks.