refactor: move optional arguments of games functions behind the ellipsis

[krlmlr]

Part of the repo-wide ellipsis migration coordinated in #2757 —
see that PR for the full rationale (CONTRIBUTING.md, Argument Order and the Ellipsis).
Based directly on main (the registry split landed via #2779); a single topic commit.

What this does

Inserts ... between the defining arguments (head) and the optional
modifiers (tail) of 36 exported games functions.
Arguments after ... become keyword-only.

• Legacy positional or abbreviated calls are recovered by the generated
  ARG_HANDLE block and emit a single lifecycle::deprecate_soft("3.0.0", …) —
  behavior is unchanged.
• No defaults change, no arguments are renamed, deprecated functions untouched.
• Registry: tools/migrations/games.R; blocks regenerated via
  Rscript tools/generate-migrations.R (idempotent, CI-checked).
• Rd usage/arguments updated mechanically (see note below).

New signatures

function	head (positional)	keyword-only tail
asym_pref()	nodes, types	type.dist.matrix, pref.matrix, loops
cit_cit_types()	n, edges, types	pref, directed, attr
cit_types()	n, edges, types	pref, directed, attr
dot_product()	vecs	directed
gnm()	n, m	directed, loops
gnp()	n, p	directed, loops
grg()	nodes, radius	torus, coords
last_cit()	n, edges	agebins, pref, directed
pa()	n, power, m	out.dist, out.seq, out.pref, zero.appeal, directed, algorithm, start.graph
pa_age()	n, pa.exp, aging.exp, m	aging.bin, out.dist, out.seq, out.pref, directed, zero.deg.appeal, zero.age.appeal, deg.coef, age.coef, time.window
pref()	nodes, types	type.dist, fixed.sizes, pref.matrix, directed, loops
sample_asym_pref()	nodes, types	type.dist.matrix, pref.matrix, loops
sample_cit_cit_types()	n, edges, types	pref, directed, attr
sample_cit_types()	n, edges, types	pref, directed, attr
sample_correlated_gnp()	old.graph, corr	p, permutation
sample_degseq()	out.deg, in.deg	method
sample_dot_product()	vecs	directed
sample_fitness()	no.of.edges, fitness.out, fitness.in	loops, multiple
sample_fitness_pl()	no.of.nodes, no.of.edges, exponent.out, exponent.in	loops, multiple, finite.size.correction
sample_forestfire()	nodes, fw.prob	bw.factor, ambs, directed
sample_gnm()	n, m	directed, loops
sample_gnp()	n, p	directed, loops
sample_grg()	nodes, radius	torus, coords
sample_k_regular()	no.of.nodes, k	directed, multiple
sample_last_cit()	n, edges	agebins, pref, directed
sample_pa()	n, power, m	out.dist, out.seq, out.pref, zero.appeal, directed, algorithm, start.graph
sample_pa_age()	n, pa.exp, aging.exp, m	aging.bin, out.dist, out.seq, out.pref, directed, zero.deg.appeal, zero.age.appeal, deg.coef, age.coef, time.window
sample_pref()	nodes, types	type.dist, fixed.sizes, pref.matrix, directed, loops
sample_sbm()	n, pref.matrix, block.sizes	directed, loops
sample_smallworld()	dim, size, nei, p	loops, multiple
sample_traits()	nodes, types, k	type.dist, pref.matrix, directed
sample_traits_callaway()	nodes, types	edge.per.step, type.dist, pref.matrix, directed
sbm()	n, pref.matrix, block.sizes	directed, loops
smallworld()	dim, size, nei, p	loops, multiple
traits()	nodes, types, k	type.dist, pref.matrix, directed
traits_callaway()	nodes, types	edge.per.step, type.dist, pref.matrix, directed

Notes for review

• The environment used to prepare this PR cannot install igraph.r2cdocs
  (GitHub API unreachable), so man/*.Rd files were updated mechanically
  (usage + ... argument item) instead of via devtools::document().
  R CMD check's usage↔formals validation passes; a follow-up
  devtools::document() run may reflow whitespace but should produce no
  semantic diff.
• Package tests that called these functions positionally were updated to
  named arguments (they are the same soft-deprecation user code will see).

[github-actions]

This is how benchmark results would change (along with a 95% confidence interval in relative change) if 6b344a7 is merged into main:

• ❗🐌as_adjacency_matrix: 747ms -> 754ms [+0.13%, +1.7%]
• ✔️as_biadjacency_matrix: 735ms -> 738ms [-1.68%, +2.56%]
• ✔️as_data_frame_both: 1.51ms -> 1.5ms [-3.07%, +1.99%]
• ✔️as_long_data_frame: 3.96ms -> 3.94ms [-3.34%, +2.6%]
• ✔️es_attr_filter: 2.66ms -> 2.67ms [-1.57%, +2.83%]
• ✔️graph_from_adjacency_matrix: 114ms -> 114ms [-1.76%, +1.58%]
• ✔️graph_from_data_frame: 3.32ms -> 3.28ms [-2.82%, +0.11%]
• ✔️vs_attr_filter: 1.52ms -> 1.51ms [-3.48%, +1.72%]
• ✔️vs_by_name: 958µs -> 956µs [-2.09%, +1.61%]
  Further explanation regarding interpretation and methodology can be found in the documentation.

[github-actions]

This is how benchmark results would change (along with a 95% confidence interval in relative change) if 782b681 is merged into main:

• ✔️as_adjacency_matrix: 801ms -> 804ms [-0.54%, +1.13%]
• ✔️as_biadjacency_matrix: 796ms -> 790ms [-2.4%, +0.94%]
• ✔️as_data_frame_both: 1.58ms -> 1.56ms [-2.78%, +0.91%]
• ✔️as_long_data_frame: 4.1ms -> 4.04ms [-4.17%, +1.1%]
• ✔️es_attr_filter: 2.86ms -> 2.82ms [-3.87%, +0.76%]
• ✔️graph_from_adjacency_matrix: 125ms -> 124ms [-2.18%, +0.84%]
• ✔️graph_from_data_frame: 3.46ms -> 3.52ms [-0.88%, +4.59%]
• ✔️vs_attr_filter: 1.66ms -> 1.65ms [-3.44%, +1.73%]
• ✔️vs_by_name: 1.06ms -> 1.06ms [-2.29%, +3.17%]
  Further explanation regarding interpretation and methodology can be found in the documentation.

[github-actions]

This is how benchmark results would change (along with a 95% confidence interval in relative change) if 5f6cd9e is merged into main:

• ✔️as_adjacency_matrix: 591ms -> 590ms [-1.58%, +1.39%]
• ✔️as_biadjacency_matrix: 566ms -> 569ms [-0.47%, +1.59%]
• ✔️as_data_frame_both: 1.24ms -> 1.23ms [-2.65%, +0.82%]
• ✔️as_long_data_frame: 3.28ms -> 3.31ms [-1.84%, +3.37%]
• ✔️es_attr_filter: 2.33ms -> 2.35ms [-1.66%, +3.04%]
• ✔️graph_from_adjacency_matrix: 97.6ms -> 99.4ms [-3%, +6.67%]
• ✔️graph_from_data_frame: 2.77ms -> 2.77ms [-1.02%, +1.4%]
• ✔️vs_attr_filter: 1.22ms -> 1.23ms [-3.31%, +5.21%]
• ✔️vs_by_name: 781µs -> 778µs [-1.6%, +0.82%]
  Further explanation regarding interpretation and methodology can be found in the documentation.