I expect that The Work of Byron Katie will be particularly useful for your type 3 classification, as it's specifically intended for getting system 1 to update on "X should/shouldn't be/do Y" beliefs. (e.g. "that person shouldn't be making munching sounds")
Per note 6, The Work actually involves a process of asking "Is that true?" about your beliefs, along with some other questions, and some pattern reversals... eg. "I shouldn't be making munching sounds", which helps in realizing that you actually have options.
For example, options that were not obvious before because system 1 was so stuck on the idea that things just shouldn't be that way. (For example, you might suddenly realize that you can wear earplugs, leave the room, politely ask someone to stop, etc.)
Of course, it might be even more helpful to question the belief "I shouldn't be petty", as it would have an even broader positive impact. ;-)
I say that because I've noticed in general that the impulses which propel people to self-improve are precisely the impulses that need to be negated in order for them to actually improve. That, e.g. a desire to "not be petty" leads precisely to a continued experience of one's self as being petty... in much the same way that the desire to not be an inadequate writer leads to a continued experience of feeling inadequate as a writer.
The thing that distinguishes these desires from healthy ones is that they're about you (the generic "you", not you, Anna Salomon specifically), rather than about the world, and that they are trying to avoid a perceived negative about the self, rather than being a desire to improve who you already are. (Even if the surface phrasing of the desire is positive, it's the emotional "tone" (as you called it) that matters.)
"Not being petty" or "being a good enough writer" are self-descriptions, not goals. A goal is, "have a good relationship with person X" or "have a good ad written". These are 1) not about one's self, and 2) can be reduced to positively-stated sensory descriptions of outside-world phenomena, without reference to your internal state.
Conversely, self-directed improvement goals (what Heidi Grant Halvorson calls "be good" goals) are negative descriptions of internal state, and lead to lots of back-and-forth and frustration because there isn't actually anything for you to optimize or move towards. All you can do is continually run headfirst into whatever you are trying to prohibit yourself from experiencing: i.e., the awareness of yourself as being "petty" or an "inadequate writer". Awareness of these self-descriptions triggers a negative self-judgment, which is painful. So your brain tries to avoid awareness, but this only perpetuates whatever outside situation is triggering the awareness (munching, needing to have a finished ad), because you're not doing anything to actually resolve the situation.
So, the solution is to question the belief that one ought not to be petty (or ought to be a good writer, or whatever), so as to discover that it is not necessary to achieve some state of perfect internal grace in order to accomplish one's true, external goals. Systems 1 and 2 will usually object, of course, because System 1 thinks that if you give up on not being petty, something awful is going to happen, and System 2 will back System 1 up with perfectly logical reasons why giving up the injunction to not be petty will in due course lead to the fall of civilization as we know it. ;-)
One of the peculiar side-effects of the way our brains render these personal injunctions is that they act like an override on both System 1 and System 2. We can't actually think about solutions to the problem of a munching noise, if the injunction is triggered by the mere thought of our not liking the noise. (Because in our mind, "not liking a noise" equates to "being petty".) So we don't even get so far as considering solutions, because we're barely even allowed to admit there's a problem.
Which is why just admitting to problems is often a helpful first step. Admitting to one's self that, "yes, actually, I am petty, if petty means disliking munching noises. And yes, I am an inadequate writer, if I define that as "not having written this ad yet"." In each case, the problem isn't the (accurate!) self-definitions, but rather, the belief that the self-definition is horrible and ought to be avoided at all costs. (And/or, the belief that the pejorative labels "petty" and "inadequate" are truly relevant or applicable to the neutral facts of the situation.)
Anyway, an awful lot of stuff is like this. Enough so that I've chosen to primarily specialize in the field of just problems that work like this. Tons of addictive and self-sabotaging behaviors build on things just like this, so I'm not going to run out of people to help any time soon. ;-)
I expect that The Work of Byron Katie will be particularly useful for your type 3 classification, as it's specifically intended for getting system 1 to update on "X should/shouldn't be/do Y" beliefs. (e.g. "that person shouldn't be making munching sounds")
Hmm... I actually read "The Work" a year or two ago, and mostly intentionally avoid recommending it to people: it seems to me that it contains powerful techniques for helping with the Type 3 classification above (as you say), but that it tends to draw people into classifyi...
Related to: Compartmentalization in epistemic and instrumental rationality; That other kind of status.
Attempted telekinesis
The case of the munching noises
The ad copy writer who doesn’t know if she’s “good enough”
Useful “telekinesis”: Separating babies from bathwater
How to distinguish?
Task type:
Type 1: Problems that System 1 can solve by itself:
Examples: Making breakfast; causing someone to know you care about them.
Suggested response: This sort of wishing is healthy, and may prompt actions that make a lot more sense than those system 2 would plan (e.g., your nonverbals as you apologize are likely to be far better if you viscerally care about your interlocutor). Leave system 1 be.
Type 2: Problems that are worth solving, but that need help from System 2:
Examples: “There’s nothing good to eat” (situation: you notice that several times, over the last hour, you’ve gone to the fridge, opened it, stared inside, closed it... and then opened it again a few minutes later -- as though to see if something good has magically materialized into the closed fridge); Feeling 'stuck' at one's job (or in a relationship); Not having enough money. (The distinguishing feature here is that system 1 has been looping on the problem for a while to no effect, and that system 2 has not yet taken a good look at the problem.)
Suggested response: Raise the problem to conscious attention; then, try to figure out what is bothering system 1; finally, decide what to do about it. As you do this, parts of the wishing will naturally shift from the general problem ("Somehow make work less stuck-feeling") to the specific strategy you've chosen ("Figure out how to renegotiate with my manager").[4]
Type 3: “Problems” that should be given up on:
Examples: “Make the munching noises go away” (in a case where you’ve decided not to); “Make San Franciscans be better drivers”; “Let me vanish into the floor.” (The distinguishing feature here is simply that these are "problems" that, on reflection, you do not wish to take action on.)
Suggested response: Find a way to let system 1 know that solving this problem isn't worth the cost, or that keeping this problem on your internal "worry/fume about" list is quite unlikely to have positive effects. For example, you might:
Examples: The problem of locating a workshop venue (during the hour at which I was trying to write the workshops ad, that October); the situation with your roommates and the dishes (while you're at work solving a coding problem).
Suggested response: Designate a particular future-you to do the task. Dialog with your "inner simulator" (your system 1 anticipations) until both system 1 and system 2 are convinced that that specific you will actually do the task, and that there is no additional positive effect to be gained via staying preoccupied now.
Type 5: Problems that System 2 needs "shower-thoughts" help with:
Examples: Archimedes' problem measuring the king's crown; "My relationship with Fred is broken, and I can't figure out what to do about it"; "How the heck can I solve that math riddle?" (The distinguishing feature here is that both: (1) the problem has already been raised to conscious attention at some point (and system 2 failed to instantly solve it); and (2) the problem is a worthy use of your shower-thoughts -- either for what it'll accomplish directly, or for the improvement it may give to your pattern of thought.))
Suggested response: This sort of wishing is healthy. Leave system 1 be.
Emotional tone:
Wishes often seem to me to have emotional tones. Some tones are simple desire (“Breakfast... mmm....”). Others have an overlayed hopelessness or bitter resignation about them (“I just always have to put up with how everyone else is incompetent”); others, still, have a tone (at least in me) of hammed-up flailing, self-pity, or desire for outside help -- as though if I just feel helpless enough, somehow a grown-up will come to the rescue ("Make the workshop crisis not be in this state... Make the workshop crisis not be in this state...").
It seems to me that it's worth installing an "alert" that sounds, in your head, whenever it hears either the hopeless/bitter/resigned tone, or the flailing/save-me tone. Both are often signs of buggy "attempted telekinesis" situations that are worth conscious debugging (a la the schema above). And the emotional tones can be easier to automatically flag.