TL;DR 1) probably not at all. 2) no. 3) imagine lots of garbage (particularly spam) that doesn't get handled nearly as fast as it should (if at all). 4) exceedingly unlikely. The network will suffer during the strike, but a lot/most of it can be cleaned up after the strike.
Update: The strike started as expected on June 5th
I'm not an employee, so I cannot answer #1 accurately. However, given the incredibly short amount of time between the initial planning phase and the start of the strike (planning started on the 30th, meaning 6 days between starting on a plan and the strike formally starting), I imagine the the answer is that they haven't prepared. Aside time, part of this boils down to the answer for #2.
As already mentioned in the comments, aside users, SOBotics, SOCVR, and Charcoal are expected to shut down, the latter of which has the single highest impact of the bunch. I don't have a source, but it has previously been stated that SO relies heavily on Charcoal for spam detection. Metasmoke, the server component of Charcoal that's also in charge of auto-flagging, went down for maintenance a couple days ago, and has already triggered questions about why spam is staying around for so long all of the sudden.2
This is with spam reports still being output to Charcoal HQ. Potentially no reports, and a significant reduction in flaggers via Charcoal (many of them are also joining the strike) is a recipe for a spam explosion, particularly when spammers notice it's down and decide to take advantage. The built-in spam protection systems are not remotely able to compete with Charcoal.
At the time of writing, there are 192 signatures, 52 of which are from mods (accounting for 9.6% of the entire moderation team1), and the remaining 140 are from curators around the network. There's several sites without a significant part of their mod team left, SO included. 14/24 active SO mods are going on strike6 (there are 27, but 3 are inactive for unrelated personal reasons), accounting for at least 80% of the work performed on SO.
There's also 3 sites fully without moderators at the time of writing, not accounting for inactive mods elsewhere in the network that may increase this number.5
All of these numbers are expected to grow over the next week, as we're still trying to get the word out to other mods. It'll also grow more during the weekdays, as some mods take time off during weekends. (It's unpaid work, it's not unreasonable to expect downtime)
For comparison, there's 14 CMs (just 14, no missing zeroes), several of which have other priorities and tasks outside moderation that can't necessarily be paused. They might be able to offset some of the striking labour, but nowhere close to prevent a work shortage, especially because they're bound by ~8 hour work days and 5 day weeks in not enough timezones to cover entire days. Even if every single CM got reassigned to network moderation, they wouldn't be able to prevent the work shortage. Mods and curators combined are generally able to cover the site 24/7, especially when it comes to spam, thanks to SmokeDetector and the accompanying cross-site cooperation, and have experience with plenty of things that CMs do not.
They might be able to source other teams that I'm unaware of, but that would also mean sourcing likely inexperienced people to offset a significantly larger amount of significantly more experienced people. It doesn't add up in their favour.
Combined with the numbers of striking curators and mods growing, they don't have any chance of offsetting the labour. The only option aside this is hiring more people, but they're likely not financially able to do so. This disregards the logistics of actually hiring people, and particularly the amount of time required for new hires to be operational on the site.
It's difficult to tell what the site will look like without moderation, but you can probably imagine lots and lots of unmoderated content floating around for longer than it should before it's handled, if it gets handled at all. Unhandled spam is likely going to be the biggest problem3, and it's also painfully visible.
A full collapse is unlikely. The network will struggle during the strike, yes, but when the strike ends, we'll be back to clean it up. The longer it goes on though, the more damage is done, some of which cannot be reversed. For the reversible damage, it will take a while to sort it out, but that's expected.
If it doesn't end, we might indeed be looking at a few collapse scenarios. I will not be speculating on what these may look like, because I like to think it won't come to that. It is exceedingly unlikely to happen due to the impact it'll have on particularly the company's reputation, and the potential for a financial impact for the company and by extension, its shareholders, who hold a lot of weight.
Also, neither side of this conflict wants a full collapse. The signals we've gotten from the company so far have been that they want to sort this out as quickly as possible, because an extended strike will have consequences for both the company and the sites in the network affected by the strike4. This doesn't guarantee any specific outcome (nor does it guarantee that certain outcomes won't happen), but it does reduce the likelihood of a collapse.
Footnotes
1: While it does account for 9.6% of the people, it does not account for 9.6% of moderation roles. Some moderators are mods on multiple sites. Though there are fewer of them than there are mods with just one diamond, they do exist, and aren't negligible. One of the more extreme examples of this is a striking moderator with 6 diamonds. See also the list of moderators grouped by user.
2: Mainly in the form of comments, though there's at least one meta post elsewhere in the network.
3: It's hard to understate the importance of Charcoal for mitigating spam. Charcoal alone is responsible for the vast majority of spam detections and detection across the network. Taking it down has an enormous impact, especially given SE's severely lacking internal spam detection systems. They've previously added system-sided blocks for spammers, but ironically, the data used to determine what metrics makes sense comes from posts caught by Charcoal's bot Smoke Detector.
4: Likely all of them, due to SD being taken down, and many curators part of the cross-site effort it's based on also going on strike. SD significantly reduces the amount of mod work and work from that site's core users required to cope with spam, which means many smaller sites are going to struggle to keep spam under control, even if no mods go on strike.
5: Mods can mark themselves as inactive, but this is not publicly displayed, so there's no way to tell without data from the company. Essentially, a site with 2 mods and one striking mod could potentially be left with 0 mods, due to the second mod being inactive for a wide array of personal reasons. There's no way to tell or guesstimate based on the number of mods either, as there isn't an expected ratio between the number of active and inactive mods. Real-life circumstances causing inactivities are unpredictable.
6: Ahead of this, many of us have also cut back on moderation activities in general, which has resulted in the flag queue on SO jumping from the 100-200 flags it has been at for the past few weeks, straight up to 1.2k. This isn't even with a full stop in moderation activities, which starts tomorrow or already, depending on the timezone. You can imagine how much work the company would have to put in to SO alone to compensate for just SO mods going on strike. This is even prior to all the other shutdowns being accounted for.
