I'm not 100% caught up on the thread, and all of this is probably pie in the sky ideas, but here it is anyways. My suggestions:
1. Introduce different chest categories (in addition to the standard chests). These different categories would have loot focused on their specific category, and perhaps items that are unique to that category. Some examples of what this could mean:
-an enchanted chest, would have a higher concentration of reagents, scrolls, spellbooks with properties (perhaps similar to the invasion spellbooks), potions, special chest-specific potions, armor/weapons that are oriented towards mage templates, etc.
-a nobleman's chest, would have higher quantities of gold, gems, special gems, jewelry, etc
-a craftsman's chest, the contents would primarily focused on resources, ingots, leather, wood, refinements, etc
-a ratman chest, bit of a wildcard chest - the ratman chest could feature odd items, random decorations that are given random, odd hues. "interesting junk" is how I would characterize it.
-an adventurer's chest, this would be the chest with a higher focus on magic armor / weapons
-a mimic, special breed of subterranean mimic that burrows into buried chests and consumes the contents - this could be something like a boss encounter that drops special loot
2. Introduce different qualities of chests. This concerns the contents mainly. I imagine there could be something like 3-5 different qualities, ranging from, for example, "rotten", "shabby", "unremarkable", "superior", "fabled". The appearance of the chest would vary depending on quality, from a crummy, broken down wooden chest to some kind of special-hued metal chest. The quality of the chest would be somewhat random - the level of the treasure map determines what sub-range you have a chance to dig up and the strength of the guardians. The quality of the magic gear would be more or less guaranteed by the quality of the chest, and the quality of the chest could also determine the quantities of stackables contained within it. If you combine this with the previous suggestion then there would be a larger variety of chests ("a superior nobleman's chest", "a shabby enchanted chest", etc). The quality of the chest would be based upon the map level and influenced by luck (RNG + luck). So if you were digging up a level 3 map for example, that might guarantee a quality of between 2-4, but the final quality is somewhat random, if that makes sense. All of this I feel would add some variety to treasure hunting.
3. This point was originally about traps and remove trap before I read the thread. Instead I'd like to offer an alternative approach to the whole skill situation. We know a chest must be found (cartography), it must be unlocked (lockpicking) and it is trapped (telekinesis, possibly remove trap). My approach to skills in this area is to give all skill levels more leeway to act, while making the high skill levels the most efficient at the task.
-Cartography: When a cartographer decodes a map, the map will show the main area where the treasure is located as it currently does, but instead of a pin, it shows a circle within which the treasure is located. The size of the circle depends on skill; the higher the skill, the smaller the area where the treasure can be found. Mining would also be rolled into cartography as has been discussed. What this means is that a lesser skilled cartographer could in theory go searching for higher level maps to some extent, but they would have to spend more time searching for it, while a skilled cartographer would be able to find it as fast as they normally would now.
-Lockpicking: Instead of the current method of lockpicking, which involves using a pick on a container and it either succeeding or failing, employ a time-based approach. A lockpicker initiates lockpicking by using the skill and targeting a container. The lockpicker then begins to work the lock - as this happens, the lockpicker uses and breaks picks, and the chance for success gradually rises the more he works the lock. Guardians will spawn throughout this process. Moving interrupts the work and getting hit has a chance of interrupting the work. If interrupted the lockpicker will be set back a bit in his work and have to use the skill on the container again but won't have to start all the way over. With this in mind, lockpickers of lower skill would be able to pick difficult locks especially if they had a group keeping guardians off of them. For a lockpicker of sufficient skill, it would only take a few seconds. For a lockpicker attempting a lock way above his skill level, maybe a few minutes and a lot of lockpicks. Kyronix, you stated earlier you've played many games with many lockpick mini-games; I'm sure among them you recall Skyrim's lockpicking game, and you might recall that any skill level can take on any lock - it's just a matter of how many lockpicks are you willing to burn through. A similar principle could apply here.
-Remove Trap: This skill should unquestionably be a part of the treasure hunting experience, but not required. It should be possible to trip the trap to access the chest's regular loot, and disarming the trap should yield a bonus to the loot. On that note, I think some chests should not have traps. Whether a chest is locked/unlocked, or has a set trap could be properties of the chest that anyone could see. As for how remove trap could possibly work, I don't really have a particularly grand idea in mind, but I think at a minimum using the skill on a trapped container should initiate a gump that shows the player the chance of success, the kind of trap it is, and a button for executing a trap removal. That way, a player would have a good idea of the risk they are about to undertake before rolling the dice with one click.
An alternative to that could be some kind of mini-game sort of like the plague beast lord surgery. I don't have a very specific way it could work in mind, but it would be timed, would require a few different tools for different parts of the mechanism, and every action taken would disturb the trap a little bit more increasing chance of failure, relative to skill level and trap difficulty. Ideally a person of the appropriate skill level should be able to remove a trap in ~10 seconds with minimal risk, while for someone of insufficient skill it would take longer and have more risk, but they could still have a chance if they understand how to navigate the puzzles. By making it a puzzle we reduce the role of pure chance in the equation but still have trap difficulty and skill being a factor in how much of a handicap the player has when navigating the trap.
4. Taking the chest with you. I think it should be possible to take a buried chest with you if you've killed all of the guardians that spawn at the beginning and the guardians that spawn during looting. Alternatively, it could also be possible to take the chest without unlocking it, however at that point it would behave much the same way as a trade quest container, where the guardian will continue to spawn at regular intervals wherever you take the chest, it cannot be recalled with, cannot go through gates, can't be put in the bank, can't be set down in a house, etc. The chest could then be unlocked/disarmed elsewhere, or it could be taken to some kind of pirate NPC at Buc's Den that buys locked treasure chests for a hefty price or a chance for a special item. At that point you wouldn't even need lockpicking or remove trap to be able to collect some kind of reward.