While trying to provide test coverage for bug 2003810, I realized that reloadBrowser test helper was misleading.
It relies on gBrowser.reloadTab, which seems to ultimately rely on WebNavigation.reload.
For some reason, this codepath seems to do very light reloads which you may not be able to do via Firefox UI!
For example, it doesn't seem to reload iframe possibly overriden location, which is really surprising.

For bug 2003810 coverage, it prevents reproducing bug 2003810 comment 0 STR.
For that STR, we ought to use sessionHistory.reload to be able to reproduce, like Firefox UI reload codepath.

So let's try to migrate our test helpers to sessionHistory.reload and fallback to browsingContext.reload.

The gBrowser.reloadTab method was doing a very light reload,
which could preserve iframe overriden locations.
The new method ensures we load the original HTML content back,
this seems to also have an impact on keep-alive which is not longer guaranteed.

Also tune the har test, now that requests are really cancelled
and we may try to fetch reponse content for such request.

https://hg.mozilla.org/mozilla-central/rev/f97aff936745