las3_pub / predictable_parallel_patterns

# Conflicts:
#	PERFORMANCE.md
#	lib/pls/CMakeLists.txt
#	lib/pls/include/pls/internal/scheduling/fork_join_task.h

This is required to correctly execute multiple, consecutive wait_for_all calls in one fork_join_sub_task.

This resulted from an merge gone wrong.

Merge: Lock Free Deque

See merge request !9

These currently do nothing but being boilerplate required to implement a new top level task.

Merge: Clean Up

See merge request !8

Merge: Cache Align thread_state

See merge request !7

The __thread implementation of thread local, static variables seems more efficient, so we activate this by default. If it is (for some reason) not available/slower than the pthread version, one can toggle it for the specific system/processor later on using macros.