proc-macro crates, it provides `momo` for de-monomorphization for now.

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

to:
 - `ThreadSafeRepository::discover_opts`
 - `ThreadSafeRepository::discover_with_environment_overrides_opts`

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

to:
 - `ThreadSafeRepository::init`
 - `ThreadSafeRepository::init_opts`

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

to:
 - `Tree<'_>::lookup_entry_by_path`
 - `Tree<'_>::peel_to_entry_by_path`

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

to:
 - `ThreadSafeRepository::open_opts`
 - `ThreadSafeRepository::open_with_environment_overrides`

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

to:
 - `Pathspec<'_>::pattern_matching_relative_path`
 - `Pathspec<'_>::is_included`

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

to:
 - `set_target_id`
 - `Platform<'_>::prefixed`

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

to `receive`

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

to:
 - `Repository::transport_options`
 - `Repository::find_object`
 - `Repository::try_find_object`
 - `Repository::find_head`
 - `Repository::try_find_head`
 - `Repository::tag`
 - `Repository::tag_reference`
 - `Repository::try_find_remote`
 - `Repository::try_find_remote_without_url_rewrite`
 - `Repository::rev_parse`
 - `Repository::worktree_stream`

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

Since the `try_find_remote_inner` is already a separate function,
there's no need to use `momo`.

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

Part of the reason is that `impl TryInto` isn't as common as `AsRef`, `AsMut` or
`Into`, but also because its error handling logic is much harder to parse and
replace in the inner function.

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

And make applying support of `gix_macros::momo` truly painless by making
it work out-of-the-box.

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

Thanks for new internal of `gix_macros::momo`!

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

So that we can slap it on more functions.

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

Now that the inner function is actually identical to the original
function before passed into `momo`, except for visibility, function name
and the lack of function doc.

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

Also, discover that apparently `momo` isn't effective where it should be
*or* the implementation is more complex.

`ty_conversions` only cover named generics, it doesn't cover
`impl Trait`.

I also changed the span for the error to `call_site`, otherwise it would
show that the `compile_error!` comes from the comment, which is
impossible.

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>

Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>