multiload: use parallel accumulators in stream-sum

[superworm]

The Makefile builds with -fno-unroll-loops. In clang this also disables loop-vectorizer interleaving, so the stream-sum reduction is vectorized with a single accumulator. Every iteration then waits on the previous vector add, which caps one thread at one 16-byte add per add latency (8 B/cycle on Neoverse N4, 16 B/cycle with 128-bit vpaddq on x86). Below DRAM sizes stream-sum was measuring this add chain, not the memory hierarchy.

Request an interleave count of 4 on the stream-sum loop so clang keeps four independent accumulators. Explicit loop hints are honored even with -fno-unroll-loops. The pragma is guarded by clang because GCC warns about unknown pragmas and the Makefile uses -Werror. GCC builds are unchanged.

Single-thread stream-sum, MiB/s, median of 3 runs (pinned, -H):

Neoverse N4 (3.6 GHz) 32k 27343 -> 91700 1m 27070 -> 74800
64m 27340 -> 52600 1g 24030 -> 25040
Xeon 6985P-C (GNR) 32k 62780 -> 124270 1m 63900 -> 93640
64m 28770 -> 28770 4g 21950 -> 21960

[gemini-code-assist]

Code Review

This pull request modifies multiload.c to add a Clang-specific loop pragma (#pragma clang loop interleave_count(4)) inside the load_stream_sum function. This change requests independent accumulators to ensure the loop remains load-bound, bypassing the limitation where -fno-unroll-loops disables Clang vectorizer interleaving. The pragma is guarded with a preprocessor check for Clang to prevent GCC compiler warnings. There are no review comments, and we have no feedback to provide.