Reland "Parameterize NavigationControllerBrowserTest etc for RenderDocument."

This is a reland of 8d6a9454

Difference is that this returns early on

All/NavigationControllerBrowserTest.SameDocumentNavigationToHttpPortZero/2

Original change's description:
> Parameterize NavigationControllerBrowserTest etc for RenderDocument.
>
> These tests are sensitive to RD's type of changes and several of them
> needed fixing already.
>
> Bug: 1100745,1068965
> Change-Id: I1bd93ebac207c502d0dda7f749a4c5f25f4e506b
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2275459
> Reviewed-by: Alex Moshchuk <alexmos@chromium.org>
> Commit-Queue: Fergal Daly <fergal@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#784274}

Bug: 1100745, 1068965
Change-Id: I4e220c7847f02a7da7326fb44cabc9cc97627a0b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2275506
Auto-Submit: Fergal Daly <fergal@chromium.org>
Commit-Queue: Alex Moshchuk <alexmos@chromium.org>
Reviewed-by: Alex Moshchuk <alexmos@chromium.org>
Cr-Commit-Position: refs/heads/master@{#784476}

testing/buildbot/filters/bfcache.content_browsertests.filter

@@ -2,13 +2,13 @@
 
 # Same-document navigation pointing to a URL that would cause a redirect if it
 # was reloaded. It never happen, because the document is restored directly.
--NavigationControllerBrowserTest.SiteInstanceChangeOnHistoryNavigation
+-All/NavigationControllerBrowserTest.SiteInstanceChangeOnHistoryNavigation/*
 
 # Expect an iframe to load again. It isn't because the page was restored from
 # the BackForwardCache.
 -BackForwardCacheMetricsBrowserTest.Features_SameOriginSubframes_CrossOriginNavigation
--NavigationControllerBrowserTest.FrameNavigationEntry_DynamicSubframeHistoryFallback
--NavigationControllerBrowserTest.FrameNavigationEntry_SubframeHistoryFallback
+-All/NavigationControllerBrowserTest.FrameNavigationEntry_DynamicSubframeHistoryFallback/*
+-All/NavigationControllerBrowserTest.FrameNavigationEntry_SubframeHistoryFallback/*
 
 # Create a URLLoaderInterceptor. It should cause the navigation to fail, but it
 # isn't using an URLLoader anymore.
@@ -23,7 +23,7 @@
 # Wait for the old process to exit. It won't, because it is still used by a
 # bfcached document.
 -DynamicIsolatedOriginTest.LockedProcessNotReusedForNonisolatedSameSiteNavigation
--NavigationControllerBrowserTest.UtilizationOfSpareRenderProcessHost
+-All/NavigationControllerBrowserTest.UtilizationOfSpareRenderProcessHost/*
 -All/RenderFrameHostManagerTest.DontSelectInvalidFiles/*
 -All/RenderFrameHostManagerTest.NavigateWithUnassignedSiteInstance/*
 -All/RenderFrameHostManagerTest.RestoreFileAccessForHistoryNavigation/*
@@ -37,7 +37,7 @@
 -All/RenderFrameHostManagerTest.RenderViewInitAfterNewProxyAndProcessKill/*
 
 # https://crbug.com/1021409
--NavigationControllerBrowserTest.BrowserInitiatedLoadPostCommitErrorPageIgnoredForFramePendingDeletion
+-All/NavigationControllerBrowserTest.BrowserInitiatedLoadPostCommitErrorPageIgnoredForFramePendingDeletion/*
 
 # Not triaged
 -All/SitePerProcessBrowserTest.ActiveSandboxFlagsRetainedAfterUnload/*