Capture "for_loop" workflow

[rauletorresc]

Context: Currently, Catalyst cannot compile and execute captured workflows with for_loop control flow. E.g.:

dev = qml.device("lightning.qubit", wires=1)

@qml.qjit(experimental_capture=True)
@qml.qnode(dev)
def circuit(n, x):

    @qml.for_loop(0, n, 1)
    def loop_rx(i, x):
            qml.RX(x, wires=0)
            return jnp.sin(x)

    # apply the for loop
    final_x = loop_rx(x)

    return qml.expval(qml.Z(0))

circuit(10, 0.3) # Array(-0.94854724, dtype=float64)

Description of the Change: Reuse the Branch interpreter created at #1468

Benefits: Extended support for plxpr capture.

• Enable pre-existent skipped test
• Update changelog
• Add more test cases

[sc-81880]

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 96.74%. Comparing base (bf7ecba) to head (207ef61).
Report is 6 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1509   +/-   ##
=======================================
  Coverage   96.73%   96.74%           
=======================================
  Files          76       76           
  Lines        8219     8236   +17     
  Branches      779      779           
=======================================
+ Hits         7951     7968   +17     
  Misses        213      213           
  Partials       55       55           

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

[PietropaoloFrisoni]

Thanks @rauletorresc!

I am not very familiar with the Catalyst pipeline, but this LGTM and I see that we are inheriting from the same PlxprInterpreter. I am very curious to test this with the features we are currently working on 🚀