test: improve test conventions

Specifically, functions that are run in the context of the test runner
are put in module `test/testutil.lua` while the functions that are run
in the context of the test session are put in
`test/functional/testnvim.lua`.

Closes https://github.com/neovim/neovim/issues/27004.

test/benchmark/autocmd_spec.lua

@@ -1,7 +1,7 @@
-local t = require('test.functional.testutil')()
+local n = require('test.functional.testnvim')()
 
-local clear = t.clear
-local exec_lua = t.exec_lua
+local clear = n.clear
+local exec_lua = n.exec_lua
 
 local N = 7500
 

test/benchmark/bench_regexp_spec.lua

@@ -1,8 +1,9 @@
 -- Test for benchmarking the RE engine.
 
-local t = require('test.functional.testutil')()
-local insert, source = t.insert, t.source
-local clear, command = t.clear, t.command
+local n = require('test.functional.testnvim')()
+
+local insert, source = n.insert, n.source
+local clear, command = n.clear, n.command
 
 -- Temporary file for gathering benchmarking results for each regexp engine.
 local result_file = 'benchmark.out'

test/benchmark/preload.lua

@@ -1,4 +1,4 @@
 -- Modules loaded here will not be cleared and reloaded by Busted.
 -- Busted started doing this to help provide more isolation.  See issue #62
 -- for more information about this.
-local t = require('test.functional.testutil')
+local n = require('test.functional.testnvim')

test/benchmark/screenpos_spec.lua

@@ -1,7 +1,9 @@
-local t = require('test.functional.testutil')()
-local api = t.api
+local t = require('test.testutil')
+local n = require('test.functional.testnvim')()
 local Screen = require('test.functional.ui.screen')
 
+local api = n.api
+
 local function rand_utf8(count, seed)
   math.randomseed(seed)
   local symbols = { 'i', 'À', 'Ⱡ', '𐀀' }
@@ -119,7 +121,7 @@ end
 
 local function benchmarks(benchmark_results)
   describe('screenpos() perf', function()
-    before_each(t.clear)
+    before_each(n.clear)
 
     -- no breakindent
     for li, lines_type in ipairs(benchmark_lines) do
@@ -134,7 +136,7 @@ local function benchmarks(benchmark_results)
           screen:attach()
           api.nvim_buf_set_lines(0, 0, 1, false, lines)
           -- for smaller screen expect (last line always different, first line same as others)
-          t.feed('G$')
+          n.feed('G$')
           screen:expect(result.screen)
           benchmark(lines, result.value)
         end)
@@ -153,9 +155,9 @@ local function benchmarks(benchmark_results)
           local screen = Screen.new(width, height + 1)
           screen:attach()
           api.nvim_buf_set_lines(0, 0, 1, false, lines)
-          t.command('set breakindent')
+          n.command('set breakindent')
           -- for smaller screen expect (last line always different, first line same as others)
-          t.feed('G$')
+          n.feed('G$')
           screen:expect(result.screen)
           benchmark(lines, result.value)
         end)

test/benchmark/treesitter_spec.lua

@@ -1,7 +1,7 @@
-local t = require('test.functional.testutil')()
+local n = require('test.functional.testnvim')()
 
-local clear = t.clear
-local exec_lua = t.exec_lua
+local clear = n.clear
+local exec_lua = n.exec_lua
 
 describe('treesitter perf', function()
   setup(function()
@@ -9,7 +9,7 @@ describe('treesitter perf', function()
   end)
 
   it('can handle large folds', function()
-    t.command 'edit ./src/nvim/eval.c'
+    n.command 'edit ./src/nvim/eval.c'
     exec_lua [[
       local parser = vim.treesitter.get_parser(0, "c", {})
       vim.treesitter.highlighter.new(parser)