Use string for SNI instead of byte[]

[twsouthwick]

This change replaces byte[] usage for SNI values to be string. There was a lot of decoding done in different areas and with this it becomes clearer that the value is a server name. The only place that actually needs the byte[] representation is the native sni - so it makes it easier to keep it as a string as much as possible.

Part of #2253

[twsouthwick]

@mdaigle here's the next step on the SSPI change

[codecov]

Codecov Report

Attention: Patch coverage is 84.34783% with 18 lines in your changes missing coverage. Please review.

Project coverage is 66.20%. Comparing base (dcf6ac4) to head (c6c95c2).
Report is 8 commits behind head on main.

Files with missing lines	Patch %	Lines
...Client/src/Interop/Windows/Sni/SniNativeWrapper.cs	70.96%	9 Missing ⚠️
...nt/netfx/src/Microsoft/Data/SqlClient/TdsParser.cs	50.00%	6 Missing ⚠️
...tcore/src/Microsoft/Data/SqlClient/SNI/SNIProxy.cs	75.00%	1 Missing ⚠️
.../netcore/src/Microsoft/Data/SqlClient/TdsParser.cs	80.00%	1 Missing ⚠️
...qlClient/src/Microsoft/Data/SqlClient/TdsParser.cs	97.50%	1 Missing ⚠️

❗ There is a different number of reports uploaded between BASE (dcf6ac4) and HEAD (c6c95c2). Click for more details.

HEAD has 1 upload less than BASE

Flag	BASE (dcf6ac4)	HEAD (c6c95c2)
addons	1	0

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2790      +/-   ##
==========================================
- Coverage   72.79%   66.20%   -6.60%     
==========================================
  Files         283      281       -2     
  Lines       59139    58847     -292     
==========================================
- Hits        43051    38960    -4091     
- Misses      16088    19887    +3799     

Flag	Coverage Δ
addons	?
netcore	69.17% <86.02%> (-6.36%)	⬇️
netfx	64.87% <83.83%> (-6.28%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

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

[paulmedynski]

I'm not familiar with this area of the code, so I reviewed for consistency and side-effects mainly.

[paulmedynski]

Thanks for addressing the naming and init comments. Just a couple more try-finally tweaks, and this will be good to go!

[paulmedynski]

Looks good, and definitely simplifies the internal APIs, reducing cognitive load.

[mdaigle]

I have one issue that I'd like addressed, then just a couple nits for naming consistency.

[mdaigle]

Looks good. Will make sure the CI passes, then merge.

[twsouthwick]

(FYI there's an auto merge option you can enable in the settings to do so when all conditions are met)