wip(assembler): round L2 -- the cross-platform hash check can fail; FP contraction changes the world

Clang -ffp-contract=on/fast and GCC's default (fast) all change the density hash on the L1 view;
contraction off reproduces L1 (= Windows S13) at W1/W12. Seed sweep: 48 FloatingIslands seeds and
14 effective SurfaceWorld seeds (34 of 48 fell on a seed-independent cube) match across Clang/GCC
and W1/W12. Standalone gains --seed/--recipe/--render-distance-chunks (viewer + headless) and a
Windows replay script. Linux standalone workers get 32 MiB pthread stacks (TLS is 17.3 MiB/thread),
so no external stack limit is needed. No Core or world-output change; MSVC compile unverified.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

Standalone/Source/StandaloneCoreScheduler.h

@@ -1,6 +1,7 @@
 #pragma once
 
 #include "VoxelForgeCore/VoxelForgeCore.h"
+#include "StandaloneWorkerThread.h"
 
 #include <algorithm>
 #include <atomic>
@@ -23,7 +24,7 @@ public:
         const std::size_t grain=std::max<std::size_t>(1,batchSize);
         std::atomic<std::size_t> next{0};
         std::vector<double> active(workers,0.0);
-        std::vector<std::thread> threads;
+        std::vector<FStandaloneWorkerThread> threads;
         threads.reserve(workers);
         for(std::size_t worker=0;worker<workers;++worker)
             threads.emplace_back([&,worker]
@@ -39,7 +40,7 @@ public:
                 active[worker]=std::chrono::duration<double>(
                     std::chrono::steady_clock::now()-started).count();
             });
-        for(std::thread& thread:threads)thread.join();
+        for(FStandaloneWorkerThread& thread:threads)thread.join();
         double total=0.0;
         for(std::size_t worker=0;worker<workers;++worker)
         {