Regarding the stuff about active versus passive voice: The difference between active and passive is that in active voice, the subject of the clause performs the action, whereas in passive voice the subject receives the action (or, to put that clause into passive voice, the subject is acted upon in a passive clause). Passive voice carries no implication that the actor is unknown, nor does it imply the agent is irrelevant. "The game was published by Activision" is an example of a passive voice sentence in which the only important information is about the actor (unless it's in answer to a question along the lines of "has the game been released yet?").
Passive voice also doesn't indicate who is doing the selling; first, second, and third person are all legal forms of passive verbs, just as they are in active verbs. Context can be used to clue you in to who is performing the action (for example, the sentence "Fred and George have been feuding recently, and Fred has been hospitalized again" carries an implication that George may have caused Fred's hospitalization, but there is doubt because such hasn't been explicitly stated), but the voice of the verb in the absence of other information will not provide any useful information about the agent. The person of the verb can carry information about who is doing the selling - I sell things while the market sells them; I buy things while the market buys them - but third person singular is the only English verb form which is normally distinct from the other verbs, and moreover the passive construction (some object or material) sells for (some amount) carries no indication of who is performing the action, as the person of the verb is determined by the subject of the sentence rather than the actor.
Regardless of how you decide to present the information, I would say that as long as it's consistent it will not be confusing for very long. Besides which, game economic systems should not allow the player to turn a profit by purchasing some item from Vendor A and then immediately selling that item back to Vendor A. Price fluctuations over time are reasonable, but that requires that you at least wait a little while before selling the item back to Vendor A, rather than selling it back before even leaving the market screen. That consideration makes me reasonably certain that whenever I see a game list two different prices for an item with one listed as the buying price and the other listed as the selling price, that the lower cost is what I get paid for an item while the higher price is what I have to pay for an item.
I also think that it might be interesting if the stability system influenced the type of market you had available - at present, the game provides something of a hybrid between a barter economy and a cash economy, where cash is the primary means of payment but goods can be provided instead. It might be interesting if that were the format for mid-stability markets, shifting towards a cash-only model in the high-stability markets and towards a full barter economy in the low-stability markets.