As far as I see, this is rather two questions. One is if companies tried to circumvent GameShark at all, as the title suggests, the other if they tried to obfuscate media content not needed as the question body suggests.
'Handling' GameShark in General
Yes, it's something done since the first flipper adds a 'TILT' light bulb. Not only did game developers try to make their games tamper proof (handlingwise or via software/hardware add ons), but even more game console makers did include measures to (at best) prevent the use of such devices in their OS/standard routines.
Obfuscating Unused Media Content
Suppose a video game company didn't have time to remove unused content in a game, [...] did any video game company try to block the use of the GameShark? Or at least make it harder for the GameShark to find hidden content?
Erm… doesn't that answer this the question already? If a company doesn't have resources/time to simply delete unused files (*1), from the generated environment before going gold, where should then the time come in to introduce complicated measures to hide this content?
Doing so would mean the developer (development tools) would have to touch especially all unused files and cloud them with some sophisticated program. Issuing a simple delete (or skip transfer) for these files seems way more sensible than writing even more software to hide them.
Now, if we drop the question requirement about unused files, but consider obscuring about every media file, the situation may look different. Still, it would be some/a lot additional development to fit in an already tight schedule (*2). Not to mention, that doing so would put the same hurdle onto the game itself, slowing it down by the additional decoding that would have had to be done (*3).
One could imagine it as, for whatever reason, planned feature to veil all content to prevent players to peek ahead. In this case, it may be as well possible that InterAct (or who ever developed it at the time) just adds the same decoder to the next release/update (*4). Adding that sounds more like a loose/loose situation for game makers who usually already struggle to get the wonderful action they had imagined worked into a real game using the limited resources of a console.
I could only assume that developers weren't too pleased about the GameShark and it's capabilities
Lets be serious, why shouldn't they? It's content they made, thus they may as well enjoy people appreciating it. But, as usual, marketing is way more relevant than any artistic part (or shame about unfinished parts). And marketing is always and everywhere quite happy about any news the game will create. A secret level… ooooh, shiney; hints about a follow up… cool; a hero selfie hidden in the game… what a great story. After all, each and every of such 'hidden' feature gets to sell more copies (*5).
but I doubt there was anything they could do about it.
Sure, there is, just, as usual, it takes quite some knowledge and even more time to implement. And at least the later is a premium during game development, isn't it?
*1 - As I understand this is about the feature of (some) GameShark versions to browse CD content for pictures, videos and sounds/music.
*2 - Basic assumption was that there is not enough time to delete unused content before going gold.
*3 - At least wrappers to decode the obfuscated/encrypted format into standard before handing it over to relevant player functions of the OS – if that's possible at all due memory and CPU constrains.
*4 - Chances would be high that such a obfuscation scheme would not be used for only one game, so there's an opportunity for even more GameShark features, making it sell better (or even twice to the same customer :))
*5 - This may destroy some players image about developers, players and the industry at all, but most 'hidden' features are not discovered by energetic hard core gamers trying every lever combination in a dungeon with differently coloured hats worn, but simply given away to journalists (or today some web forum) at some point to increase sales.