I think the performance and security level is the same between those two because they use the same level encryption, we only change the order. Is my reasoning right?
Yes, the performance is exactly the same, and I can't see why Veracrypt would even offer a choice, as I can't see any security implications in changing the order in which the ciphers are applied. Maybe someone on crypto.stackexchange.com could offer a reason why I'm wrong, though.
I'd go further and say that triple encryption, while it certainly protects against there being catastrophic failures in two of the three ciphers, is paranoid. I'd only consider this worthwile if I needed to directly hand over my encrypted hard drive to an attacker and leave him alone with it for the next 20 years, in which all kinds of breakthroughts might be made.
I can't say much about Whirlpool vs Streebog, except that just because the NSA touches something doesn't mean it's containing a backdoor. The SHA-2 family has been reviewed by lots of renowned cryptographers who have found it solid and while that doesn't prove anything, withstanding analysis by experts is really the only measure of quality we have in the field of cryptography.
There are some attacks against both Whirlpool and Streebog, but the best ones are still completely impractical for real-world attacks (AFAIK). I think the best attack on reduced-round Whirlpool can create a collision with a time complexity of 2^120. That's still out of reach. The best known attack against Streebog has an even higher time complexity. There are also attacks against reduced-round SHA-2 hash functions, but it's difficult to put these 3 hash functions in any order security-wise.
Finally, while I'm not entirely sure what the hash function is used for in Veracrypt (besides key derivation), a weakness in the hash function should not affect the security of the encryption if you use a strong passphrase.
2) I did a container on my usb stick. Does encryption use more storage than non-encrypted files
No, it doesn not use more storage (except for a very small header that doesn't exist with unencrypted partitions).
Veracrypt encrypts disk blocks, not files. Each encrypted block is exactly the same size as the unencrypted block. It has to be; otherwise the encrypted partition would not fit into the space reserved for the partition on disk.