Add more shim representation tests

[t0yv0]

This PR simply adds test cases around tfgen and shim representations of various schemata found in SDKv2 and PF. For PF I went through the docs to cover each major advertised case. What's of special interest is pinning down when Pulumi generates object types.

It seems to be happening in two cases:

1. when TF specifies object types directly in PF:
   - Single Nested Attribute
   - Object Attribute
   - Single Nested Block

2. when TF specifies a block with MaxItems=1 that is a list or set in TF but is flattened to an object in Pulumi:
   - In SDKv2 this is typically a list-nested block with MaxItems=1, very common
   - In PF this may be a block with List Validators indicating that at most 1 item is supported; seems rare

It seems helpful to pin down what shim.Schema captures in each of these cases.

[t0yv0]

Guess could be interesting to add for PF set-of-object and list-of-object cases for attributes.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 56.90%. Comparing base (65dd45d) to head (35e0c00).
Report is 2 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2456      +/-   ##
==========================================
- Coverage   56.90%   56.90%   -0.01%     
==========================================
  Files         369      369              
  Lines       50733    50736       +3     
==========================================
+ Hits        28869    28870       +1     
- Misses      20285    20288       +3     
+ Partials     1579     1578       -1     

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

[t0yv0]

Hmm linter is happy locally but not in CI. Perhaps not aligned on linter versions.

 ::high file=pkg/tfshim/sdk-v2/cty.go,line=164,col=32::G115: integer overflow conversion uint -> int64 (gosec)
  
  level=warning msg="[config_reader] The output format `github-actions` is deprecated, please use `colored-line-number`"

[VenelinMartinov]

LGTM.

Linking #2231 as it covered some of the same grounds and had some interesting observations on similarities/differences between various TF Types represented in the Shim layer.

[pulumi-bot]

This PR has been shipped in release v3.92.0.