Whether or not there is a bounty, I think the port should be of bitcoin-wallet. If there is a bounty, I think there should be a separate one for bitcoinj. I think the currency and name operations should be separate bounties. I think the name operations would be significantly more difficult than currency operations, because there is no existing base for the name transactions. So I think the bounty split should reflect that.
Regarding the person who posted on Reddit, it looks like he
charges as low as 2 BTC. Now, I assume that would only be for currency operations, and that name operations would cost significantly more.
So there are some options. One, pay this guy. Two, see if Namecoin gets accepted to GSoC and see if a student wants to do the Android wallet project. Three, have someone from within the Namecoin devs that has the time do the port. Fourth option is to put up a bounty that is less than 2 BTC and hope that someone that is qualified, but willing to accept less than 2 BTC, comes along.
In my opinion it would be best to try option two first before considering paying for the wallet. I probably would have tried porting bitcoin-wallet myself, except I wanted to save something for the GSoC that is fast approaching. I already developed Armory and the Android port of NMControl to the point where they are probably no longer suitable GSoC projects (though name operations in Armory may still be suitable as a GSoC project (I'm not really sure how long it would take)).