[Nir-Az]

Overview: Cherry-pick of #15330 to the 2.58.3 release branch — in the viewer, a post-processing filter control (e.g. decimation magnitude, rotation) applied the effect but the control snapped back to its previous value; it now stays at the value the user set.

Tracked on [RSDEV-12488], [RSDEV-12502]

Why

Regression from PR #15029 (async UI option writes) present in 2.58.3. Software post-processing filters emit no on_options_changed echo, so the control's cached value was never refreshed after the write and reverted once the ~2 s user-request mask expired (the effect applied, but the slider snapped back).

Summary

• Cherry-pick of viewer: fix post-processing filter controls reverting to default value #15330 (merge cf8648a26), clean apply, no conflicts.
• Software post-processing filter options write synchronously (value read back → no revert); FW/sensor options stay async.
• Includes the review follow-ups: reset last_user_set_stopwatch on sync writes, scope the invalidate gate to a local error buffer, and a viewer UI regression test.

Test plan

• Verified on development: viewer UI test post_processing_value_persists passes on D455; confirmed live in the viewer; ran in nightly (gui+nightly context).
• CI on this cherry-pick PR.

---

🤖 Generated by AI

[sysrsbuild-gh-agentic]

[rs-agentic bot] Automated review of cherry-pick #15345 to r/2.58.3.

The fix is sound: routing software-filter writes through a synchronous path with immediate readback directly addresses the root cause (no on_options_changed echo -> stale cache -> revert). Overall the approach is correct. Several items flagged below — see inline comments.