feat: support HTTP&gRPC&pg set timezone

[Taylor-lagrange]

I hereby agree to the terms of the GreptimeDB CLA

What's changed and what's your intention?

Support set/get timezone in HTTP&gRPC&pg&mysql

Use unify SQL layer to dispose SQL SET time_zone = xxx and show variables xxxx, make all protocol support set/get timezone

HTTP

Add a Header named x-greptime-timezone, user can carry timezone by this http header

gRPC

Add timezone in GreptimeRequest::RequestHeader make gRPC query aware timezone info

Related pr GreptimeTeam/greptime-proto#132

PG

Use SHOW VARIABLES system_timezone; and SHOW VARIABLES timezone; to query system timezone and timezone

Use Set time_zone = xxx to set user timezone, system time indicate by timezone in frontend config file

MYSQL

Because MYSQL federated intercept SQL parse like SHOW VARIABLES xxx.

In mysql, we still use Select @@system_timezone; and Select @@timezone; to query system timezone and timezone

Use Set time_zone = xxx to set user timezone, same with PG

Documents will be updated after #2907 finish

Checklist

• I have written the necessary rustdoc comments.
• I have added the necessary unit tests and integration tests.
• This PR does not require documentation updates.

Refer to a related PR or issue link (optional)

#2907

[codecov]

Codecov Report

Attention: 32 lines in your changes are missing coverage. Please review.

Comparison is base (1fc168b) 85.52% compared to head (69f9e68) 84.93%.
Report is 23 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3125      +/-   ##
==========================================
- Coverage   85.52%   84.93%   -0.59%     
==========================================
  Files         818      825       +7     
  Lines      133850   135081    +1231     
==========================================
+ Hits       114474   114730     +256     
- Misses      19376    20351     +975     

[killme2008]

LGTM

[waynexia]

I'm curious about the change from "".to_string() to String::default(). Is clippy complaining about it?

[Taylor-lagrange]

In my practice, "".to_string() to String::default() may help to discover case like

fn s(s: Option<String>) -> String {
    s.unwrap_or("".to_string())
}

clippy won't complain about it, but String::default() will suggust use unwrap_or_default