A bundle of tweaks

[NiLuJe]

• InputContainer: Drop dead code in the form of seqtext & doc fields. I'm not sure if they're remnants of KPV or of an aoborted design, but it's been very dead code for a very long while.
  This spurred a bit of cleanup along the way ;p.
• Persist: Make sure data is flushed to disk on save (fix Support for the Kobo Clara 2E (goldfinch) #9552 (comment))
• GestureDetector: Only log adjusted gestures for gestures actually requiring a rotation adjustment ;).
• Kobo: Flag all the boards similar to the Libra 2 as broken^Wbuggy, and enable the full set of workarounds there (re: Support for the Kobo Clara 2E (goldfinch) #9552).
• USBMS: Add a nicer info-level log message when entering USBMS, because apparently people are confused by the UIManager:quit ones.
• Kobo: Downgrade a few non-failure PM log messages to debug.
• Kindle: Ditto (that one's for you @hius07 ;p).
• GH: Update the issue template to mention that toggling verbose debug logging requires a restart.
• AutoSuspend: Update the standby help message to be friendlier to everyone, but specifically warn affected users on the aforementioned broken^Wbuggy boards ;).
• Input: Allow disabling rotation_map entirely, and do so on the PB700 (fix 2022.8 page turn buttons need to be reversed when orientation is set to "horizontal" #9556)
• PocketBook: Move low-level input event type translation to the low-level input modules.
• OTM: Deal with a fun little quirk of the defaults.lua migration (fix 2022.10 crashes if DCREREADER_CONFIG_MARGIN_SIZES present in defaults.persistent.lua #9700)
• Kindle: Make sure Device:untar actually works when targeting the userstore fuse proxy (fix can't seem to look up words with chinese dictionary #9704)

Depends on koreader/koreader-base#1544 & koreader/koreader-base#1545

---

This change is 

[NiLuJe]

(Rebase me).

First commit ended up sliiightly scarier than I expected, but... 'tis the season :D.

[poire-z]

(Approved as "ok on the principle", not that I approve the unit tests failing :)

[NiLuJe]

Yeah, unit tests should be fixed by the base bump ;p.

[hius07]

Side question. Should we use such alignment more often? (looks good)

koreader/frontend/ui/widget/container/scrollablecontainer.lua

Lines 82 to 90 in 8921e80

	self.ges_events = {
	ScrollableTouch = not ignore.touch and { GestureRange:new{ ges = "touch", range = range } } or nil,
	ScrollableSwipe = not ignore.swipe and { GestureRange:new{ ges = "swipe", range = range } } or nil,
	ScrollableHold = not ignore.hold and { GestureRange:new{ ges = "hold", range = range } } or nil,
	ScrollableHoldPan = not ignore.hold_pan and { GestureRange:new{ ges = "hold_pan", range = range } } or nil,
	ScrollableHoldRelease = not ignore.hold_release and { GestureRange:new{ ges = "hold_release", range = range } } or nil,
	ScrollablePan = not ignore.pan and { GestureRange:new{ ges = "pan", range = range } } or nil,
	ScrollablePanRelease = not ignore.pan_release and { GestureRange:new{ ges = "pan_release", range = range } } or nil,
	}

[NiLuJe]

I mean, I'd be all for it, as that how I setup clang-format for C stuff ;o).

Here, the pattern kinda screamed for it, though, so it was an easy decision (by which I mean, it's a very neat series ;)).

[NiLuJe]

Reviewed 79 of 84 files at r1, 4 of 4 files at r3, 2 of 2 files at r4, 7 of 7 files at r5, all commit messages.
Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @Frenzie, @pazos, and @poire-z)

[NiLuJe]

Reviewed 6 of 6 files at r6, all commit messages.
Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @Frenzie, @pazos, and @poire-z)

[NiLuJe]

Reviewed all commit messages.
Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @Frenzie, @pazos, and @poire-z)

[yparitcher]

@NiLuJe
Regression

The 4 places you changed self.ges_events from {} to nil do not let me do any gestures in the reader (even swipe to open the menu)

./luajit: frontend/ui/widget/container/inputcontainer.lua:257: bad argument #1 to 'pairs' (table expected, got nil)

[hius07]

(Writing here as it's one of the latest PR changing the Menu widget)

On a tap the coroutine is created:

koreader/frontend/ui/widget/menu.lua

Lines 487 to 498 in 068809b

	function MenuItem:onTapSelect(arg, ges)
	-- Abort if the menu hasn't been painted yet.
	if not self[1].dimen then return end
	local pos = self:getGesPosition(ges)
	if G_reader_settings:isFalse("flash_ui") then
	logger.dbg("creating coroutine for menu select")
	local co = coroutine.create(function()
	self.menu:onMenuSelect(self.table, pos)
	end)
	coroutine.resume(co)
	else

This was introduced in #1637 to support OPDS authorization.
Since then OPDSBrowser has been changed a lot and now does not check for a coroutine existence.
So is this coroutine creation needed anymore?

[poire-z]

Not really familiar with corountines (or what I learned about them when I needed to make Trapper, I have forgotten), but indeed, we never really need to use coroutines, so I guess if we'd rewrite this menu select code now, we wouldn't even think about adding anything coroutine in there :)
So, if everything keeps working without, nothing against cleaning that out.

[Frenzie]

I forgot that was in there. What that code is doing seems sensible for that kind of scenario though? What happens now? I don't really use OPDS much, let alone with authorization.

[hius07]

What happens now?

Earlier there was an option to authorize to opds server manually via the dialog entering login/pwd.
Now login/pwd must be saved to the server properties, no online dialog for login exists.

[Frenzie]

If it's not used anywhere I suppose it should be fine to remove since then it's also somewhat untested after all. :-)