perf: import pgraph without matplotlib, spatialmath or scipy

[petercorke]

import pgraph took 0.62 s. Profiling showed almost all of it was two imports that are only needed to plot:

• from spatialmath.base.graphics import axes_logic (the one thing pgraph uses from spatialmath): 0.4 s, because import spatialmath brings in sympy, scipy, PIL and, today, matplotlib; and
• matplotlib.pyplot, mpl_toolkits.mplot3d and friends, which two module-level classes (_Arrow3D, _Loop3D) needed just to be defined.

Making matplotlib lazy alone would save nothing, since spatialmath loads it too; both have to go.

Change

• _loop_path, _Arrow3D and _Loop3D move to a new private module pgraph/_plot.py.
• Graph._edge_artist imports from it, and plot, highlight_vertex, highlight_edge and highlight_path import pyplot (and plot the axes_logic of spatialmath), when called.
• No behaviour change. The one test import moves (from pgraph._plot import _Arrow3D, _Loop3D).

import pgraph	before	after
time	0.62 s	0.04 s (numpy alone is 0.03 s)
matplotlib, spatialmath, scipy loaded	yes	no

tests/test_lazy_imports.py starts a fresh interpreter to check what an import, a build, three planners, the spatial queries, subgraph, copy, the matrices, export and a pickle round trip load (nothing heavy), and that plotting in 2D and 3D (arrows, loops, highlight_path) imports matplotlib at that point.

762 tests pass, mypy clean, docs build with no warnings.

This does not wait for the spatialmath lazy-import work: it is independent of it, and the saving is the same with or without it.

🤖 Generated with Claude Code

[codacy-production]

Up to standards ✅

🟢 Issues 0 issues

Results:
0 new issues

View in Codacy

🟢 Metrics 14 complexity · 0 duplication

Metric	Results
Complexity	14
Duplication	0

View in Codacy

_{NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.28%. Comparing base (c9d38e6) to head (7759a58).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #70      +/-   ##
==========================================
+ Coverage   97.26%   97.28%   +0.01%     
==========================================
  Files           3        4       +1     
  Lines        1570     1581      +11     
==========================================
+ Hits         1527     1538      +11     
  Misses         43       43              

☔ View full report in Codecov by Harness.
📢 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.