Engine: drop the layer-loop memory flush in preset M

The ten network layer dispatches only need dispatch serialization, not a
global memoryBarrier flush: each layer's cross-block traffic goes through
its own buffers, already published by the preceding stage's flush. An
empty compute->compute pipeline barrier keeps the ordering and removes
the flush cost: whole-frame median 2515.0 -> 2431.7 us at Balanced
(1488x837 -> 2560x1440), 3818.9 -> 3773.3 at Quality, 1892.6 -> 1810.6
at Performance, byte-identical to the captured launches (chain test
10/10 layers 100.0%, frame test inf dB, 8- and 16-frame outputs match).

Also documents this session's discarded probes (intra-frame barrier
relaxation, colSum subgroup shuffle, downsample BLOCK_Y 32, expansion
EO stride 24, cross-frame pipelining analysis, the NO_ROUND 7.0%
rounding-free ceiling) and adds the round_half.glsl include guard.

packaging/RELEASE_NOTES.md

@@ -88,6 +88,7 @@ This release adds an optional accuracy mode and fixes the GLIBC compatibility fa
 - **More useful failure logs.** Rejected input/output formats and dimensions are now recorded even after startup, with repeated errors rate-limited. Output-wait CPU and wall time are reported separately, and GPU timing queries are deferred until asynchronous work finishes.
 - **More conservative accuracy texture builds.** Strict surface conversions use separate helper functions, avoiding a compiler failure when the original denormal handling is retained.
 - **Accuracy build safeguards.** The builder refuses to label a directory containing older fast binaries as an accuracy set. Configuration tests cover every release target, missing-set fallback, invalid markers, environment precedence, and the default-off behavior.
+- **Preset M network dispatch barriers.** The ten layer dispatches no longer flush global memory between layers; an empty pipeline barrier keeps the serialization the layers actually need. Captured-frame replays stayed byte-identical at every stage. On an RX 9070 XT at Balanced (1488×837 → 2560×1440) the network median dropped 1863.0 → 1765.6 µs and the whole frame 2515.0 → 2404.6 µs; Quality and Performance improved 36.6 and 62.8 µs with byte-identical outputs.
 
 ## Also included from 0.1.2