[Trivial]std.uni: Fix a -dip1000 compilable issue; appeared since #6041

[carblue]

Using posix.mak from #6278
make -f posix.mak std/uni.test
results in:
...

T=`mktemp -d /tmp/.dmd-run-test.XXXXXX` &&                                                              \
  (                                                                                                     \
    ../dmd/generated/linux/release/64/dmd -od$T -conf= -I../druntime/import  -w -de -dip25 -m64 -fPIC -transition=complex -O -release -dip1000  -main -unittest generated/linux/release/64/libphobos2.a -defaultlib= -debuglib= -L-ldl -cov -run std/uni.d ;     \
    RET=$? ; rm -rf $T ; exit $RET                                                                   \
  )
std/uni.d(7414): Error: scope variable chars assigned to non-scope parameter inp calling std.uni.Grapheme.opOpAssign!("~", const(dchar)[]).opOpAssign
std/uni.d(8363): Error: template instance `std.uni.Grapheme.__ctor!dchar` error instantiating
std/uni.d(7414): Error: scope variable chars assigned to non-scope parameter inp calling std.uni.Grapheme.opOpAssign!("~", const(int)[]).opOpAssign
std/uni.d(8371): Error: template instance `std.uni.Grapheme.__ctor!int` error instantiating

sources used (on Linux x86_64):
dmd : Latest commit 17bc56c
druntime: Latest commit f2711a3
phobos : Latest commit 6947d80

[DmitryOlshansky]

Seems harmless. But this is strange though as a template should likely infer scope-ness

[carblue]

Yes, I agree about inferring scope for params in templates and it used to be inferred (and probably will be again). Walter is heavily working on safe/scope issues currently, thus this issue may be only temporary; but it's not the only location where I observed changed safe/scope/dip1000 behavior recently.

[wilzbach]

Yeah the compiler should be smarter :/