extending getSymbols(.rda/.csv) to work with tick data

Currently, setSymbolLookup does not use user defined col.names, only allowing OHLCV(A). It would be useful to extend to other financial data formats.

```

AUDUSD <- structure(c(0.98646, 0.98652, 0.98647, 0.98644, 0.98639,
0.98635, 0.98664, 0.98669, 0.98665, 0.98661, 0.98658, 0.98653),
.Dim = c(6L, 2L), .Dimnames = list(NULL, c("AUDUSD.bid", "AUDUSD.ask")),
index = structure(c(1288573200.756, 1288573200.699, 1288573201.500,
1288573201.999, 1288573202.103, 1288573202.479), tzone = "",
tclass = c("POSIXct", "POSIXt")), .indexCLASS = c("POSIXct", "POSIXt"),
tclass = c("POSIXct", "POSIXt"), .indexTZ = "", tzone = "",
class = c("xts", "zoo"))
save(AUDUSD, file="AUDUSD.rda")
require(quantmod)
setSymbolLookup(AUDUSD = list(src="rda", dir=".", extension="rda",
col.names=c("bid", "ask")))
getSymbols("AUDUSD")
# Does not work
getSymbols("AUDUSD", col.names = c("bid", "ask"))
# Works
```

Start with the setSymbolLookup and getSymbols entry points described in the issue, tracing how rda and csv sources handle col.names. Compare the configured lookup path with the direct getSymbols call, then verify that user-defined bid and ask names work for tick data without breaking the existing OHLCV(A) behavior.

Written by the indexing model from the issue text.