Add stack-usage stats

See merge request !338

add Context out of bounds write test

See merge request !336

EMPER's Runtime now installs the SIGSEGV handler that checks if the
segmentation fault was caused by hitting the guard page. If the guard
page was hit, then the handler prints an error message.

Also make ContextOutOfBoundsTest more robust by not creating an
infinite recursion, which may be optmized out by the compiler.

[Emper.hpp] Remove unnecessary 'static' keyword

See merge request !337

Add an SIGSEGV handler to the ContextOutOfBoundsTest to ensure that
the received SIGSEGV is caused by accessing the guard page.

Add emper::printInfo()

See merge request !333

The test is a simple infinite recursion hitting the guard page.
The new guard-page-test-runner skips the test if emper::STACK_GUARD_PAGE
is false.

[Context] Add option for guard page at the bottom of stack

See merge request !335

[meson] fix context_alignas option

See merge request !334

The used variables are named assume_*.
And meson does not support concatenating different variable types using
'+' use string.format instead.

[Context] Make context size configurable

See merge request !331

Extend README and switch to one line per sentence

See merge request !329

Also keep the context size at 64 KiB (there was a comment that
errorneously indicated that the context size is 4 MiB).

do not warn about missing atomic::{wait,notify}

See merge request !332

This allows emper to be build with --meson-fatal-warnings on in our i4lab.

I do not see a reason to do so any more (there probably never was).

Use constexpr for memory manager in context manager

See merge request !330

[build] Check if std::atomic wait()/notify{_all}() is avaialble

See merge request !326

use emper_fibril in continuation-stealing tests

See merge request !328

Add support for continuation stealing

See merge request !327

Thanks to Nicolas Pfeiffer for writing the initial prototypical
implementation of continuation stealing and the cactus stack
mechanism, on which this is based.

Co-authored-by: Nicolas Pfeiffer <pfeiffer@cs.fau.de>

Fixes builds on Debian stable, which currently fail with

ccache c++ -Ieval/io_latency.p -Ieval -I../eval -Iemper -I../emper -Iemper/io -I../emper/io -Iemper/include -I../emper/include -I/usr/include -fdiagnostics-color=always -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -Wnon-virtual-dtor -Wextra -Wpedantic -Werror -std=c++2a -O2 -g -DBOOST_ALL_NO_LIB -pthread -isystem../subprojects/liburing-liburing-2.0/src/include -isystemsubprojects/liburing-liburing-2.0/src/include -MD -MQ eval/io_latency.p/IoLatency.cpp.o -MF eval/io_latency.p/IoLatency.cpp.o.d -o eval/io_latency.p/IoLatency.cpp.o -c ../eval/IoLatency.cpp../eval/IoLatency.cpp: In function ‘void watchFunc()’:
../eval/IoLatency.cpp:45:17: error: ‘struct std::atomic<bool>’ has no member named ‘wait’
   45 |   readSubmitted.wait(false, std::memory_order_acquire);
      |                 ^~~~

Introduce AbstractFiber as superclass of all scheduled control flows

See merge request !325

It is safe to perform a relaxed load in isRunnable(), since the
information is potentially outdated immediately after it was looked
at.

Also, since Fiber::isRunnable() is a virtual method, keeping its
definition in the header has no real advantage, hence we move it out
of the header.

[io/Stats] don't try to print worker stats when using a single uring

See merge request !324