Org reader: fix parsers relying on parseFromString

Emphasis was not parsed when it followed directly after some block types
(e.g., lists).

The org reader uses a wrapper for the `parseFromString` function to
handle org-specific state. The last position of a character allowed
before emphasis was reset incorrectly in this wrapper. Emphasized text
was not recognized when placed directly behind a block which the reader
parses using `parseFromString`.

Fixes: #4784

test/Tests/Readers/Org/Block/List.hs

@@ -243,4 +243,15 @@ tests =
       mconcat [ para "CLOSED: [2015-10-19 Mon 15:03]"
               , bulletList [ plain "Note taken on [2015-10-19 Mon 13:24]" ]
               ]
+
+  , "Markup after header and list" =:
+      T.unlines [ "* headline"
+                , "- list"
+                , ""
+                , "~variable name~"
+                ] =?>
+      mconcat [ headerWith ("headline", [], []) 1 "headline"
+              , bulletList [ plain "list" ]
+              , para (code "variable name")
+              ]
   ]