Re-add skip_on_cran() to sim_pop_data_multi test; sharpen NEWS bullet

skip_on_os("linux") was correctly removed by the cross-platform fix
(quantize_t_next ensures consistent RNG consumption everywhere).
skip_on_cran() is restored because the test grid (5 lambdas x 2 sample
sizes x nclus=10 parallel simulations) is too slow for CRAN's per-check
time budget.

Also replaces the vague "Removed lingering terminology discrepancies"
NEWS bullet with a specific description of the strat_ests_barplot() and
strat_ests_scatterplot() axis-label corrections.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01B5FrE8RZ6EjFiz7q8zqbfg

Author: claude

NEWS.md

@@ -28,7 +28,9 @@
   shifted the random-number stream out of sync and made simulated values, and
   their snapshots, differ between macOS, Windows, and Linux). Simulated
   values change slightly as a result of this fix. (#447)
-* Removed lingering terminology discrepancies
+* Corrected default axis labels in `strat_ests_barplot()` (`xlab`) and
+  `strat_ests_scatterplot()` (`ylab`) to say "seroincidence" rather than
+  "seroconversion"/"incidence".
 * `load_noise_params()` and `load_sr_params()` now fail gracefully with informative messages when internet resources are unavailable, complying with CRAN policy (#505)
 * Added Version Crosswalk article to pkgdown website to help users migrate code from v1.3.0 to v1.4.0
   - Provides clear tables comparing old and new function names

tests/testthat/test-sim_pop_data_multi.R

@@ -1,4 +1,5 @@
 test_that("`sim_pop_data_multi()` works consistently", {
+  skip_on_cran()
   # Load curve parameters
   dmcmc <- typhoid_curves_nostrat_100