Remove psyir if branch

[hbrunie]

This PR adds a new PsyIR transformation.

What this transformation should do exactly should be discussed between @schreiberx @arporter and other interested.

Currently there is no test, it is only reusing the code present in Martin's PR #2801.

[codecov]

Codecov Report

Attention: Patch coverage is 24.56140% with 43 lines in your changes missing coverage. Please review.

Project coverage is 99.80%. Comparing base (f534847) to head (d0aa7ca).

Files with missing lines	Patch %	Lines
...lone/psyir/transformations/remove_ifblock_trans.py	24.56%	43 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2869      +/-   ##
==========================================
- Coverage   99.88%   99.80%   -0.09%     
==========================================
  Files         359      360       +1     
  Lines       50860    50917      +57     
==========================================
+ Hits        50804    50818      +14     
- Misses         56       99      +43     

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

[arporter]

This part would need to either be removed entirely (and the transformation purely on PSyIR information for determining compile-time constants) or replaced with functionality that enables a list of names/symbols to be provided to the apply method (to support run-time constants).

[arporter]

Thanks for this Hugo. Obviously this will need full testing etc. Also, the reading of the json file needs to be replaced with one or more parameters to apply. One of us could take this on if need be?

[schreiberx]

@arporter One year passes really quickly.
If it's not too late, and if one of you still wants to take care of it, please go ahead.
If not, please let me know.

Background information: This would be a good feature for, e.g., NEMO & Croco to eliminate if-branches once the parameters are known to be constant over runtime.