Is the possibility of discovering private keys with random testing by considering the increasing number of e-wallet?
Yes, there is a tiny tiny tiny possibility of generating private keys that were generated before. The chances are described on the above posts, the fact that there is 1 in x possibilities shouldn't trick you. We can't even pretend to understand how big is that number, it's above human's standards. It's never happened and I hope it'll never be.
what the prevention solution in cryptocurrency such as bitcoin is?
You can't prevent someone from calculating. Everyone's free to brute force bitcoin addresses.
Is there any mechanism such as "I am not a Robot" (Captcha) or blocking the user with the iterative wrong private key?
In order for this to happen Bitcoin addresses must be generated over a centralized authority that will filter users' private keys to prevent any collisions. I guess you understand that this is against decentralization, against its purpose.
I believe that you haven't been convinced that it's impossible to achieve such thing, so I'll tell you one last thing:
With the current difficulty, in order to mine a block, you'll have to try among ~23,100,000,000,000 * 2
32 = ~99,213,744,537,600,000,000,000 combinations. If you'd be able to generate 1 hash, then you'd have 1 in that number, to be rewarded with 6.25 fresh bitcoins.
Compare these numbers:
- 99,213,744,537,600,000,000,000
- 1,461,501,637,330,902,918,203,684,832,716,283,019,655,932,542,976
Not to mention that generating an address requires more computational power than performing an SHA256d. A p2pkh requires you to perform ECDSA, SHA256, RIPEMD-160, SHA256 again, SHA256 again and then just a base58 encoding.