dialects: (seq) Add CompRegOp

[Moxinilian]

This op allows modeling registers. This is for now a very basic implementation which only partially matches the CompRegOp upstream. But it should be enough for many practical uses already!

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 89.68%. Comparing base (df1c825) to head (a7642fa).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2404   +/-   ##
=======================================
  Coverage   89.68%   89.68%           
=======================================
  Files         346      346           
  Lines       41418    41430   +12     
  Branches     6175     6176    +1     
=======================================
+ Hits        37144    37156   +12     
  Misses       3372     3372           
  Partials      902      902           

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

[lucjaulmes]

I think maybe you could add the optional $powerOnValue operand? Otherwise LGTM 🚀

Optional attributes inner_sym and name will need #2407.

(Though name is a bit of a weird one, it seems to be inferred from “the SSA name of the operation’s first result” and to only be printed if not matching with the SSA name. I think just adding it as an optional attribute down the line should be enough?)

[Moxinilian]

I think maybe you could add the optional $powerOnValue operand? Otherwise LGTM 🚀

It's a bit silly but adding the power on value triggers a bug in the assembly format thing that is a bit over my head to attempt to fix. Perhaps we can do without for now?

[lucjaulmes]

Sure. Maybe file a bug about the double-optionals? We’ll have to revisit compregop anyway.

[luisacicolini]

lgtm!

[Moxinilian]

#2413