* test/src/treesit-tests.el (treesit-query-node-only-and-grouped):
Remove unused variables.

Before the fix, if user uses the NODE-ONLY option, we don't keep
the capture names in the results, then predicates won't work
because they can't reference capture names.

* src/treesit.c (query_capture_remove_capture_name): New
function.
(Ftreesit_query_capture): Use the new function to remove capture
names AFTER running the predicate.
* test/src/treesit-tests.el:
(treesit-query-node-only-and-grouped): New test.

This introduces the predicate frame-initial-p, which uses
struct frame.output_method or struct terminal.type to detect
initial_terminal without relying on its name (bug#80629).
For some prior discussion, see:
https://lists.gnu.org/r/emacs-devel/2019-12/msg00480.html
https://lists.gnu.org/r/emacs-devel/2020-01/msg00120.html

* doc/lispref/frames.texi (Frames): Document frame-initial-p.
(Finding All Frames): Fix grammar.
* etc/NEWS (Lisp Changes in Emacs 31.1): Announce frame-initial-p.
* lisp/desktop.el (desktop--check-dont-save):
* lisp/emacs-lisp/debug.el (debug):
* lisp/frameset.el (frameset-restore):
* lisp/menu-bar.el (menu-bar-update-buffers):
* lisp/xt-mouse.el (turn-on-xterm-mouse-tracking-on-terminal):
Use frame-initial-p instead of checking the "initial_terminal" name.
* lisp/emacs-lisp/byte-opt.el: Mark frame-initial-p as error-free.

* src/pgtkterm.c (pgtk_focus_changed): Use IS_DAEMON in place of
Fdaemonp, thus also accepting a named daemon session.
* src/terminal.c (decode_tty_terminal): Clarify commentary.
(Fframe_initial_p): New function.
(syms_of_terminal): Expose it.
(init_initial_terminal): Update commentary now that
menu-bar-update-buffers uses frame-initial-p (bug#53740).

* test/lisp/xt-mouse-tests.el (with-xterm-mouse-mode): Simulate the
lack of an initial terminal by overriding frame-initial-p now
that turn-on-xterm-mouse-tracking-on-terminal uses it.
* test/src/terminal-tests.el: New file.

* lisp/emacs-lisp/comp.el (comp--func-unique-in-cu-p): Check
only the callee name instead of requiring global uniqueness.
* test/src/comp-resources/comp-test-direct-call.el: New file.
* test/src/comp-resources/comp-test-direct-call-dup.el: New file.
* test/src/comp-tests.el (comp-tests-has-direct-call-p)
(comp-tests-direct-call-with-lambdas)
(comp-tests-direct-call-with-duplicate-names): New tests.

* lisp/emacs-lisp/comp.el (comp--add-cond-cstrs): Skip
negated equality assumptions only for relational non-immediate
operands. Keep immediate false-branch constraints.
* test/src/comp-resources/comp-test-funcs.el
(comp-test-80327-hash, comp-test-80327-f): Add reproducer.
* test/src/comp-tests.el (comp-tests-bug-80327): Add regression
test.

* test/lisp/vc/vc-tests/vc-tests.el
(vc-test--run-maybe-unsupported-function): Remove dummy branch in
`condition-case`.
* test/lisp/arc-mode-tests.el (define-arc-mode-test-on-type):
Use `error-slot-value`.

* test/src/process-tests.el (process-tests--ignore-EMFILE):
* test/src/filelock-tests.el (filelock-tests-unlock-spoiled)
(filelock-tests-kill-buffer-spoiled):
* test/lisp/vc/vc-git-tests.el (vc-git-test--run):
* test/lisp/proced-tests.el (proced--assert-process-valid-cpu-refinement):
* test/lisp/emacs-lisp/pp-tests.el (pp-tests--sanity):
* test/lisp/emacs-lisp/edebug-tests.el (edebug-tests-with-normal-env)
(edebug-tests-post-command):
* test/lisp/autorevert-tests.el (auto-revert--deftest-remote):
* test/infra/android/test-controller.el (ats-connect):

* test/src/thread-tests.el (threads-join-error):
* test/src/emacs-module-resources/mod-test.c (Fmod_test_signal):
* src/print.c (print_bind_overrides):
* lisp/url/url-future.el (url-future-finish, url-future-errored)
(url-future-call, url-future-cancel): Make sure signal's data is a list.

* test/src/emacs-module-tests.el (mod-test-non-local-exit-signal-test):
Adjust accordingly.

This reverts commit feac53141577161c32a7a6dfe75399a5ae98a7c1.

This patch has deactivated 253 test cases without a sufficient reasoning.
Instead it speaks about a shotgun in its commite message.

The patch is reverted because

- It hasn't been discussed on emacs-devel. It should have, because it is
  a serious change in our infrastructure (new official tag :nobatch). Any
  documentation of this change, for example in test/README, is missing.

- The proper way to deactivate such tests would have been

  (skip-when noninteractive)

  Even better to skip for the respective reasons.

- There is no fault report. There is no information about how these tests
  have failed. Since it hasn't been a problem so far for us, nobody will
  work on a fix forever.