The tree sitter elm parser contains a c symbol which collides with other
linked symbols. This PR downgrades the tree sitter elm parser to a
version which doesn't have this problem.

Release Notes:
- Fixed crash when parsing elm files

In https://github.com/zed-industries/zed/pull/2790 I added an extra drag
event on mouse_up which signaled the end of a drag event, as mouse_up
event themselves wouldn't reliably fire if users moved their mouse too
quickly. This broke the assumptions of the terminal element. This PR
adds filters to all current on_drag handlers which removes this new
event.

Release Notes:

- Fixed a bug causing terminal links to never open (preview only)
- Fixed a bug in terminal link detection causing it to miss files with a
`-` in it

Parallelize Vector Database calls for project semantic search.

Release Notes: (Preview-only)

- Parallelize Vector database calls for project semantic search. Cuts
query time by 2/3rds.
- Removed default keymap for old semantic search modal.

Updated app icon and icon system updates

This PR cascades the split resizing to adjacent splits, if the current
split has already hit the minimum size. This PR also adds support for
detecting the end of a drag event to GPUI, via a bool on the dispatched
drag.

Release Notes:

- Made split resizing more flexible

Move semantic search from navigation modal, to project search option.
This PR is intended to be released in Preview only, and requires an
opt-in semantic_index option to enable. Without this opt-in setting
enable, the user should perceive no differences between previous project
search.

Release Notes: (Preview-only)

- Added Semantic Search as a opt-in feature within Project Search
- Show indexing feedback on indexing process within project search view

Release Notes:

- vim: add a mode indicator
([#409](https://github.com/zed-industries/community/issues/409))

Now updated screenshots with @iamnbutler 
<img width="1043" alt="Screenshot 2023-07-25 at 11 11 57"
src="https://github.com/zed-industries/zed/assets/94272/8301479a-8b58-42d8-81a1-bc40e1e0a4df">
<img width="1043" alt="Screenshot 2023-07-25 at 11 12 00"
src="https://github.com/zed-industries/zed/assets/94272/89c3b8bd-9cbc-4fd7-ad10-dac5538ed3a3">
<img width="1043" alt="Screenshot 2023-07-25 at 11 12 12"
src="https://github.com/zed-industries/zed/assets/94272/adc87fe3-a720-4779-853b-df9443407046">

I want to use this on another branch, but it's a sweeping change,
so this prepares the ground for it. This can always be reverted if
it doesn't work out.

This PR adds command palette actions for:

- `sort lines case sensitive`
- `sort lines case insensitive`
- `reverse lines`
- `shuffle lines`

Closes out:

- https://github.com/zed-industries/community/issues/658

and partially closing out:

- https://github.com/zed-industries/community/issues/57 (which is
currently a top-ranked issue)

There are issues with dedupe lines and I didn't try to tackle converting
variable names between different conventions. I'll likely close out 57
with a note to just upvote the remaining individual issues

Release Notes:

- added command palette actions for `sort lines case sensitive`, `sort
lines case insensitive`, `reverse lines`, and`shuffle lines`
(([#57](https://github.com/zed-industries/community/issues/57)),
([#658](https://github.com/zed-industries/community/issues/658)))

Co-Authored-By: Mikayla Maki <mikayla.c.maki@gmail.com>

This adds tree-sitter grammars for both Elm and GLSL, with injections
for GLSL embedded within Elm. It also adds an `outline.scm` for Elm,
though limitations in the tree-sitter grammar meant that I wasn't able
to get it looking exactly how I'd have liked.

In particular, it wasn't clear how to nicely annotate functions in the
outline as being functions, or how to prevent the fields of a record
declaration from being increasingly indented.


![image](https://github.com/zed-industries/zed/assets/285821/544bc00b-3bfc-4ec7-be9d-764b9f0292ab)

![image](https://github.com/zed-industries/zed/assets/285821/74e57e95-bf87-4989-ae29-a2f625141bcf)

![image](https://github.com/zed-industries/zed/assets/285821/9e283c78-66d5-4c15-9827-1b5b446cdc37)

fixes https://github.com/zed-industries/community/issues/598

Release Notes:
- Added syntax highlighting for the Elm and GLSL languages

Co-Authored-By: Mikayla Maki <mikayla.c.maki@gmail.com>

Co-Authored-By: Nate Butler <iamnbutler@gmail.com>

Co-Authored-By: Julia <30666851+ForLoveOfCats@users.noreply.github.com>

@mikayla-maki for 👀 

[This PR added in the ability to rename a file via
`enter`](https://github.com/zed-industries/zed/pull/2784). Previously,
`enter` was used to both open a file and confirm a rename, so this PR
changes the opening of a file to use `space`, which is what VS Code
uses. It also makes a bit more sense because now `enter` is just used to
start a rename and confirm the rename, vs being used for 2 different
actions.

N/A on the release notes, as I adjusted the release note in the
previously-tagged PR.

Release Notes:

- N/A

Follow-up to https://github.com/zed-industries/zed/pull/2777
Refs https://github.com/zed-industries/community/issues/1770

In this PR, I reworked the way that git statuses are retrieved. In a
huge repository like `WebKit`, the really slow part of computing a list
of git statuses is the *unstaged* portion of the diff. For the *staged*
diff, `git` can avoid comparing the contents of unchanged directories,
because the index contains hashes of every tree. But for the *unstaged*
portion, Git needs to compare every file in the worktree against the
index. In the common case, when there are no changes, it's enough to
check the `mtime` of every file (because the index stores the mtimes of
files when they are added). But this still requires an `lstat` call to
retrieve each file's metadata.

I realized that this is redundant work, because the worktree is
*already* calling `lstat` on every file, and caching their metadata. So
in this PR, I've changed the `Repository` API so that there are separate
methods for retrieving a file's *staged* and *unstaged* statuses. The
*staged* statuses are retrieved in one giant batch, like before, to
reduce our git calls (which also have an inherent cost). But the
`unstaged` statuses are retrieved one-by-one, after we load files'
mtimes. Often, all that's required is an index lookup, and an mtime
comparison.

With this optimization, it once again becomes pretty responsive to open
`WebKit` or `chromium` in Zed.

Release Notes:

- Optimized the loading of project file when working in very large git
repositories