fix(wasm): drop pthreads — fixes UI hang during prefill and generation (#32)

pthreads + ASYNCIFY conflict: emscripten_sleep() on the main thread
doesn't yield properly when pthreads are active (Atomics.wait-based
scheduling interferes with ASYNCIFY's stack unwinding). Result: UI
hangs during both prefill and generation despite yield calls.

Fix: remove -pthread and PTHREAD_POOL_SIZE. Keep ASYNCIFY + SIMD128
+ relaxed-simd + O3 + LTO. The emscripten_sleep(0) calls in both
prefill (every 2 tokens) and generation (every 4 tokens) now work
correctly, keeping the browser responsive.

quant.h's matmul still uses pthread_create internally, but without
-pthread they become no-ops in Emscripten — single-threaded matmul
with SIMD vectorization.

Binary: 384K → 320K (-17%, no pthread runtime).

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: unamedkr

wasm/build.sh

@@ -1,12 +1,12 @@
 #!/bin/bash
 # Build quant.cpp WASM demo
-# pthreads for parallel matmul + SIMD + ASYNCIFY for UI streaming
+# SIMD + ASYNCIFY for streaming (no pthreads — conflicts with ASYNCIFY sleep)
 set -e
 
 SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
 PROJECT_DIR="$(dirname "$SCRIPT_DIR")"
 
-echo "=== Building quant.cpp WASM (pthreads + SIMD) ==="
+echo "=== Building quant.cpp WASM (SIMD + ASYNCIFY) ==="
 
 if ! command -v emcc &>/dev/null; then
     echo "Error: emcc not found. Install Emscripten SDK."
@@ -22,24 +22,21 @@ emcc "$SCRIPT_DIR/quant_wasm.c" \
     -msimd128 \
     -mrelaxed-simd \
     -flto \
-    -pthread \
     -s WASM=1 \
-    -s INITIAL_MEMORY=256MB \
-    -s MAXIMUM_MEMORY=4GB \
     -s ALLOW_MEMORY_GROWTH=1 \
+    -s MAXIMUM_MEMORY=4GB \
+    -s INITIAL_MEMORY=256MB \
     -s EXPORTED_FUNCTIONS='["_main","_wasm_load_model","_wasm_generate","_wasm_generate_async","_wasm_model_info","_wasm_is_ready","_malloc","_free"]' \
     -s EXPORTED_RUNTIME_METHODS='["UTF8ToString","allocateUTF8","FS"]' \
     -s FORCE_FILESYSTEM=1 \
     -s MODULARIZE=0 \
-    -s ENVIRONMENT='web,worker' \
+    -s ENVIRONMENT=web \
     -s NO_EXIT_RUNTIME=1 \
     -s ASSERTIONS=0 \
     -s STACK_SIZE=1MB \
     -s ASYNCIFY \
     -s 'ASYNCIFY_IMPORTS=["emscripten_sleep"]' \
     -s ASYNCIFY_STACK_SIZE=65536 \
-    -s PTHREAD_POOL_SIZE=4 \
-    -s PTHREAD_POOL_SIZE_STRICT=0 \
     -lm \
     -DNDEBUG \
     -D__EMSCRIPTEN__ \