fix: consistent message hashes

[skosito]

https://github.com/zeta-chain/protocol-private/issues/181

I don't think there is danger of hash colisions, since this is hashed and signed by tss in zetaclient. However, there is good point to have hashes organized a bit better, and with expected field size.

Changes:

• add ZETACHAIN at begining of hash (i think this is optional, but i guess nice to have)
• order of arguments like instruction identifier, chain id, nonce is same in every hash
• for instruction names, i guess there are a lot of strategies to have them occupy consistent space and not use variable length strings, i used enum, but open for discussion

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 0% with 10 lines in your changes missing coverage. Please review.

Project coverage is 2.85%. Comparing base (cf38502) to head (fe5e506).

Files with missing lines	Patch %	Lines
programs/gateway/src/lib.rs	0.00%	10 Missing ⚠️

Additional details and impacted files

@@                       Coverage Diff                        @@
##           non-security-related-findings     #72      +/-   ##
================================================================
- Coverage                           2.86%   2.85%   -0.02%     
================================================================
  Files                                  1       1              
  Lines                                453     456       +3     
================================================================
  Hits                                  13      13              
- Misses                               440     443       +3     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[lumtis]

LGTM