Extend single mode symplectic to act on multiple modes 📏

[sduquemesa]

Context: No method is provided to extend a single-mode symplectic to act on several modes.

Description of the Change:

• Implements the new extend function to do just so.
• It works on sparse matrices as well. COO, DIA and BSR sparse matrices will be casted to CSR.

Benefits:

• New useful functionality 📏
• Works on sparse matrices.

Possible Drawbacks: None

Related GitHub Issues: None

[codecov[bot]]

Codecov Report

Merging #347 (59d8725) into master (d8def78) will not change coverage. The diff coverage is 100.00%.

@@            Coverage Diff            @@
##            master      #347   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           24        24           
  Lines         1725      1737   +12     
=========================================
+ Hits          1725      1737   +12     

Impacted Files	Coverage Δ
thewalrus/symplectic.py	100.00% <100.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update d8def78...59d8725. Read the comment docs.

[nquesada]

so this acts the same symplectic in all the modes?

[sduquemesa]

Thanks @ilan-tz and @nquesada,

so this acts the same symplectic in all the modes?

@nquesada Correct!

---

As per @ilan-tz comments:

In the current implementation, I believe extend operates in the xpxp convention rather than the default xxpp convention.

Resolved ✅

When I input a single mode as an argument to extend, I get a ValueError -- however, perhaps it should reproduce expand in this case, allowing me to act just on the mode I specify?

I've merged the two functions so you should see the same expected behaviour.

Currently, the entries corresponding to the modes on which the symplectic doesn't act are assumed to be 0 -- however, they should be the identity.

My bad, sorry 🙈 Resolved. 

Perhaps we should add this into expand after all, rather than have a separate function? What do you think?

Agree!