phase 0+1: project scaffold + xserv-cuda crate

- Cargo workspace with xserv-cuda crate
- CUDA FFI bindings (cudart: memory, stream, device, error)
- GpuBuffer RAII wrapper with H2D/D2H/D2D copy
- CudaStream wrapper with RAII Drop
- CachingAllocator with size-bucketed free lists
- PinnedBuffer for page-locked host memory
- Device info query via cudaDeviceGetAttribute
- Vector-add CUDA kernel smoke test
- Integration test suite (11 tests)
- build.rs: cc crate compiles .cu for SM 12.0
- sync-and-build.sh for remote build on dash5
- Roadmap doc (docs/00-roadmap.md) and Phase 0+1 design doc

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

crates/xserv-cuda/src/stream.rs

@@ -0,0 +1,33 @@
+use crate::error::{self, Result};
+use crate::ffi;
+
+pub struct CudaStream {
+    raw: ffi::CudaStream,
+}
+
+impl CudaStream {
+    pub fn new() -> Result<Self> {
+        let mut raw = std::ptr::null_mut();
+        error::check(unsafe { ffi::cudaStreamCreate(&mut raw) })?;
+        Ok(Self { raw })
+    }
+
+    pub fn synchronize(&self) -> Result<()> {
+        error::check(unsafe { ffi::cudaStreamSynchronize(self.raw) })
+    }
+
+    pub fn as_raw(&self) -> ffi::CudaStream {
+        self.raw
+    }
+}
+
+impl Drop for CudaStream {
+    fn drop(&mut self) {
+        if !self.raw.is_null() {
+            unsafe { ffi::cudaStreamDestroy(self.raw) };
+        }
+    }
+}
+
+// Can move across threads, but not shared without synchronization
+unsafe impl Send for CudaStream {}