fix: eval context fixes and new error types

[rgrassian-split]

Signed-off-by: Robert Grassian [email protected]

This PR

• Changes the evaluation context merge to create and return a new object rather than modify itself. This fixes a bug where the openfeature api level context was being overwritten by the client's context
• Creates TargetingKeyMissingError and InvalidContextError, along with their error codes
• Adds an evaluation context setter on the client

[codecov-commenter]

Codecov Report

Merging #43 (9da38aa) into main (df0c033) will decrease coverage by 0.81%.
The diff coverage is 78.57%.

@@            Coverage Diff             @@
##             main      #43      +/-   ##
==========================================
- Coverage   93.59%   92.78%   -0.82%     
==========================================
  Files          17       17              
  Lines         281      291      +10     
==========================================
+ Hits          263      270       +7     
- Misses         18       21       +3     

Flag	Coverage Δ
unittests	92.78% <78.57%> (-0.82%)	⬇️

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

Impacted Files	Coverage Δ
open_feature/exception/exceptions.py	78.26% <66.66%> (-4.10%)	⬇️
open_feature/open_feature_client.py	95.77% <66.66%> (-1.33%)	⬇️
...n_feature/evaluation_context/evaluation_context.py	90.00% <100.00%> (ø)
open_feature/flag_evaluation/error_code.py	100.00% <100.00%> (ø)

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[matthewelwell]

Thanks @rgrassian-split, totally agree with the new approach for the context merging, thanks for submitting this. I've added some comments in a few places.

[beeme1mr]

@matthewelwell @ajhelsby any concerns about merging this?