Use single https session to retrieve all rock-on definitions #2707

[phillxnet]

By sharing a single https session, instantiated on the first GET, we avoid major cumulative delays and reduce client/server resource overhead. Note that we still employ a dedicated https session to retrieve our 'index' file of 'root.json'.

Includes

• Minor fstring adoption in modified area of code.
• Info log of definitions retrieval and initial parsing time.
  Intended as a diagnostic aid, and to assist with field feedback on this process of indeterminate duration.
• Individual file retrieval timeout reduced to 5 seconds.

Fixes #2707

[phillxnet]

Testing

In the associated issue there are a number of comment details the apparent improvement here, and a freshly built RPM (15.5 amd64 build host) with this patch in-place installs successfully and is able to acquire the initial Rock-on list, and then update twice there-after with the following new info log entries detailing the retrieval time:

[17/Oct/2023 17:37:13] INFO [storageadmin.views.rockon:505] Rock-on definitions retrieved in: 5.99 seconds.
[17/Oct/2023 17:37:33] INFO [storageadmin.views.rockon:505] Rock-on definitions retrieved in: 5.98 seconds.
[17/Oct/2023 17:37:49] INFO [storageadmin.views.rockon:505] Rock-on definitions retrieved in: 6.01 seconds.

With an overall API (/api/rockons/update) time similarly logged as follows.
N.B. we also have a newer gunicorn version using a more appropriate worker type in play here: see #2706

127.0.0.1 - - [17/Oct/2023:17:37:16 +0100] "POST /api/rockons/update HTTP/1.0" 200 0 "https://rleap15-5.lan/" "Mozilla/5.0 (X11; Linux x86_64; rv:109.0) Gecko/20100101 Firefox/115.0" 9441ms
127.0.0.1 - - [17/Oct/2023:17:37:36 +0100] "POST /api/rockons/update HTTP/1.0" 200 0 "https://rleap15-5.lan/" "Mozilla/5.0 (X11; Linux x86_64; rv:109.0) Gecko/20100101 Firefox/115.0" 9235ms
127.0.0.1 - - [17/Oct/2023:17:37:52 +0100] "POST /api/rockons/update HTTP/1.0" 200 0 "https://rleap15-5.lan/" "Mozilla/5.0 (X11; Linux x86_64; rv:109.0) Gecko/20100101 Firefox/115.0" 9244ms

[phillxnet]

@Hooverdan96 Thanks for linking in that operationally orientated issue. Much appreciated. I created a new issue as this may not be the whole story, but I'm confident this and the referenced gunicorn update/worker change should at least help. Hopefully representing a fix for our occasionally referenced time-outs.

@FroggyFlox & @Hooverdan96 I'll go ahead and get this merged ready for our nearing next testing rpm release.