Cherry-pick gtest fix for asan tests.

[eugenis]

Summary

https://github.com/google/googletest/commit/681454dae48f109abf68c424c9d2e6db9a092238

Clone+exec death test allocates a single page of stack to run chdir + exec on.
This is not enough when gtest is built with ASan and run on particular
hardware.

With ASan on x86_64, ExecDeathTestChildMain has frame size of 1728 bytes.

Call to chdir() in ExecDeathTestChildMain ends up in
_dl_runtime_resolve_xsavec, which attempts to save register state on the stack;
according to cpuid(0xd) XSAVE register save area size is 2568 on my machine.

This results in something like this in all death tests:

Result: died but not with expected error.
...
[  DEATH   ] AddressSanitizer:DEADLYSIGNAL
[  DEATH   ] =================================================================
[  DEATH   ] ==178637==ERROR: AddressSanitizer: stack-overflow on address ...

PiperOrigin-RevId: 278709790

[eugenis]

ping

[pcc]

LGTM

llvm/utils/unittest/googletest/src/gtest-death-test.cc

GTEST_DEATH_TEST_CHECK_SYSCALL_(sigaction(
       SIGPROF, &ignore_sigprof_action, &saved_sigprof_action));
 #   endif  // GTEST_OS_LINUX
 
 #   if GTEST_HAS_CLONE
   const bool use_fork = GTEST_FLAG(death_test_use_fork);
 
   if (!use_fork) {
     static const bool stack_grows_down = StackGrowsDown();
-    const size_t stack_size = getpagesize();
+    const size_t stack_size = getpagesize() * 2;
     // MMAP_ANONYMOUS is not defined on Mac, so we use MAP_ANON instead.
     void* const stack = mmap(NULL, stack_size, PROT_READ | PROT_WRITE,
                              MAP_ANON | MAP_PRIVATE, -1, 0);
     GTEST_DEATH_TEST_CHECK_(stack != MAP_FAILED);
 
     // Maximum stack alignment in bytes:  For a downward-growing stack, this
     // amount is subtracted from size of the stack space to get an address
     // that is within the stack space and is aligned on all systems we care