Migrate off BrowserThread::DeleteSoon in /content/browser/renderer_host

Instead use base::DeleteSoon which enforces usage of TaskTraits.

This CL is a no-op as-is. It was recently discovered however that some
callers did BrowserThread::DeleteSoon() with pending tasks running on
different task queues (different traits  -- e.g.

TaskTraits to make this more obvious. Please review whether calls in
this CL can be migrated as-is or need additional traits to match
potentially pending tasks.

Split from https://chromium-review.googlesource.com/c/chromium/src/+/1894109
for cursory review.

This CL was uploaded by git cl split.

R=kbr@chromium.org


TaskPriority: :BEST_EFFORT) which can result in out-of-order deletion...
BrowserThread: :DeleteSoon() is being migrated to base::DeleteSoon() w/
Bug: 1019767
Change-Id: I2bfb718aa066e09050197565d36f8f11b34e8a93
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1894882
Auto-Submit: Gabriel Charette <gab@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Kenneth Russell <kbr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#711805}

content/browser/renderer_host/render_process_host_impl.cc

@@ -464,9 +464,8 @@ class SessionStorageHolder : public base::SupportsUserData::Data {
   ~SessionStorageHolder() override {
     // Its important to delete the map on the IO thread to avoid deleting
     // the underlying namespaces prior to processing ipcs referring to them.
-    BrowserThread::DeleteSoon(
-        BrowserThread::IO, FROM_HERE,
-        session_storage_namespaces_awaiting_close_.release());
+    base::DeleteSoon(FROM_HERE, {BrowserThread::IO},
+                     session_storage_namespaces_awaiting_close_.release());
   }
 
   void Hold(const SessionStorageNamespaceMap& sessions, int widget_route_id) {