[Frontend] Make AOT happen for any function annotated with AbstractValue signatures

[erick-xanadu]

Context: In preparation for quantum subroutines, I found the need to compile the following function:

@catalyst.qjit(target="mlir")
def identity(x : AbstractQreg()):
    return x
print(identity.mlir)

However, one small impediment was this check for types to decide whether AOT compilation is possible or not.

Description of the Change: Generalize small check to decide whether AOT compilation is possible or not by relying on isinstance of AbstractValue over ShapedArray. ShapedArray is derived from AbstractValue

Benefits: More general code, small progress towards quantum subroutines.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 96.60%. Comparing base (f484b02) to head (dc42cb8).
Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1774   +/-   ##
=======================================
  Coverage   96.60%   96.60%           
=======================================
  Files          82       82           
  Lines        9211     9212    +1     
  Branches      872      872           
=======================================
+ Hits         8898     8899    +1     
  Misses        254      254           
  Partials       59       59           

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[dime10]

Are you planing on using the full end-to-end compilation on subroutines?

[erick-xanadu]

@dime10 at the moment, I just want to be able to generate plxpr / jaxpr. And stitch it together inside qjit during tracing.