build(deps): add libstdcxx pin, bump CondaPkg

[cjdoris]

Add libstdcxx to CondaPkg.toml with version "<=julia" to improve libstdc++ compatibility with the Julia toolchain. Bump CondaPkg compat to 0.2.30 to leverage related fixes. Update release notes to reflect the improved compatibility.

[thchr]

For anyone potentially hitting the issue I hit just now and tracing it to here: I believe pulling in this update in my main environment (i.e. up-to-date PythonCall/CondaPkg), but not another project-specific environment (i.e. v0.9.26 PythonCall/v0.2.29 CondaPkg), produced the following error:

julia> using [some local package involving a CondaPkg.toml file]
    CondaPkg Found dependencies: /home/tchr/.julia/packages/PythonCall/avYrV/CondaPkg.toml
    CondaPkg Found dependencies: /mnt/c/Users/tchr/[some local folder]/julia/SymmetryBases/CondaPkg.toml
    CondaPkg Found dependencies: /home/tchr/.julia/environments/v1.11/CondaPkg.toml
    CondaPkg Found dependencies: /mnt/c/Users/tchr/[some local folder]/julia/MPBUtils/CondaPkg.toml
    CondaPkg Found dependencies: /home/tchr/.julia/packages/PythonCall/IOKTD/CondaPkg.toml
    CondaPkg Resolving changes
             + libstdcxx
             + libstdcxx-ng
             + nodefinder (pip)
             + numpy
             + openssl
             + phasemap (pip)
             + pymeep
             + pynormaliz (pip)
             + python
             + scipy
             + uv
    CondaPkg Initialising pixi
             │ /home/tchr/.julia/artifacts/cefba4912c2b400756d043a2563ef77a0088866b/bin/pixi
             │ init
             │ --format pixi
             └ /mnt/c/Users/tchr/[some local folder]/julia/BandGraphs/test/.CondaPkg
✔ Created /mnt/c/Users/tchr/[some local folder]/julia/BandGraphs/test/.CondaPkg/pixi.toml
    CondaPkg Wrote /mnt/c/Users/tchr/[some local folder]/julia/BandGraphs/test/.CondaPkg/pixi.toml
             │ [dependencies]
             │ openssl = ">=3, <3.6, >=3, <3.6"
             │ libstdcxx = "<=julia"
             │ uv = ">=0.4"
             │ libstdcxx-ng = ">=3.4,<14.0, >=3.4,<14.0"
             │ pymeep = "*"
             │ numpy = "*"
             │ scipy = "*"
             │
             │     [dependencies.python]
             │     channel = "conda-forge"
             │     build = "*cpython*"
             │     version = ">=3.9,<4, >=3.8,<4"
             │
             │ [project]
             │ name = ".CondaPkg"
             │ platforms = ["linux-64"]
             │ channels = ["conda-forge"]
             │ channel-priority = "strict"
             │ description = "automatically generated by CondaPkg.jl"
             │
             │ [pypi-dependencies]
             │ phasemap = "*"
             │ nodefinder = "*"
             └ pynormaliz = "*"
    CondaPkg Installing packages
             │ /home/tchr/.julia/artifacts/cefba4912c2b400756d043a2563ef77a0088866b/bin/pixi
             │ install
             └ --manifest-path /mnt/c/Users/tchr/[some local folder]/julia/BandGraphs/test/.CondaPkg/pixi.toml
Error:
  × failed to solve the conda requirements of 'default' 'linux-64'
  ╰─▶ Cannot solve the request because of: No candidates were found for libstdcxx <=julia.


ERROR: InitError: failed process: Process(setenv(`/home/tchr/.julia/artifacts/cefba4912c2b400756d043a2563ef77a0088866b/bin/pixi install --manifest-path '/mnt/c/Users/tchr/[some local folder]/julia/BandGraphs/test/.CondaPkg/pixi.toml'`, [...............]

The error was resolved by updating my out-of-date project environment to the latest PythonCall.jl version - but a bit tricky that one environment's PythonCall/CondaPkg versions could somehow influence another's.