revert(build): drop 4.6.0 as a primary R version option

#183 added it so the per-minor pass could run under 4.6. That was
unnecessary: build-all-versions already loops every installed interpreter
and runs build-all.R --sensitive-only for each non-primary minor, so the
4.6 pass happens when the pipeline runs with the slot's normal
R_VERSION.

Worse than unnecessary, it is a footgun. R_VERSION selects the *primary*
minor, whose build lands in the generic slot. Choosing 4.6.0 for a slot
whose generic binaries are 4.5-built would publish 4.6 binaries there and
break every 4.5 client - the mirror image of the bug this all started
with.

.crow/build-all-versions.yaml

@@ -35,14 +35,16 @@ variables:
       - 'resolute'
     default: '3.24'
   R_VERSION:
+    # The slot's *primary* R minor: what `local/build-all.R` builds into the
+    # generic slot. The loop below already runs `--sensitive-only` for every
+    # other installed minor, so filling a non-primary minor's gap needs this
+    # left alone, not changed.
+    #
+    # 4.6.0 was briefly offered here (#183) and removed: selecting it for a
+    # slot whose generic binaries are 4.5-built would publish 4.6 binaries
+    # into the generic slot and break every 4.5 client.
     description: 'Primary R version under /opt/R.'
-    # The supported window is latest plus the two previous, which the build
-    # images install as R_VERSION_LATEST/PREV1/PREV2. 4.6.0 was missing here,
-    # so no pipeline could run the per-minor pass for it and its slots kept a
-    # backlog: rlang exists for 4.4 and 4.5 on amd64/resolute but not 4.6,
-    # which is how an R 4.6.1 client ended up loading a 4.5.3 binary.
     options:
-      - 4.6.0
       - 4.5.3
       - 4.4.3
     default: 4.5.3